CommandCam is a simple and easy to use command line webcam image grabber for Windows. It captures a single image from a webcam and stores it in a bitmap file. CommandCam is ideal for use in batch files and other situations where you want a very simple way to automate image capture. CommandCam uses Microsoft’s DirectShow API to access webcams, so it should work with most USB cameras.
CommandCam (the source code and the executable version) are published under the GNU General Public License (version 3).
Download CommandCam
CommandCam can be downloaded from my page on Github:
https://github.com/tedburke/CommandCam
The executable file, CommandCam.exe, can be downloaded directly using the following link:
CommandCam.exe (65KB, date: 21-Oct-2021)
The full CommandCam source code, which is contained in a single C++ file, is also available:
CommandCam.cpp (17KB, date: 21-4-2012)
Running CommandCam
To run CommandCam:
- Firstly, open a console window. From the Windows start menu, select Start -> Accessories -> Command Prompt.
- Make sure your webcam is plugged in.
- Move to the directory where you downloaded CommandCam.exe. For example, cd "My Documents\Downloads".
- Type CommandCam and press return.
- The captured image will be saved to the file image.bmp.
Here’s how that appears in the console window.
The above example captured the following image of my cup of tea to a file called "image.bmp" (the image was converted to PNG format prior to uploading).
CommandCam Options
Several useful options can be specified using command line arguments.
To specify a time delay between the camera being turned on and the image being grabbed, use the "/delay" option to specify a time delay in milliseconds. For example, to add a 5 second delay,
CommandCam /delay 5000
The default output filename for the grabbed image is "image.bmp". To specify a different filename, use the "/filename" option. If the filename contains any spaces, enclose it in inverted commas. For example, to save the image as "face.bmp":
CommandCam /filename face.bmp
By default, the first available video capture device will be used. To specify a particular device (if you have more than one), use the "/devnum" option. For example, to open the second video capture device:
CommandCam /devnum 2
Alternatively, to specify a capture device by name, use the "/devname" option. If the device name contains any spaces, it should be enclosed in inverted commas. For example, to select a device called “USB Video Device”:
CommandCam /devname "USB Video Device"
To specify a capture device by serial number (for example if you wish to select between two cameras of the same model), use the "/devserial" option. For example, to select a device with the serial number “314159265”:
CommandCam /devserial 314159265
By default, CommandCam does not display any video on the screen before capturing the image. However, you can enable a video preview window using the "/preview" option.
CommandCam /preview
To list the available capture devices, for example to check which device number corresponds to which device, use the "/devlist" option.
CommandCam /devlist
To list the available capture devices with extra detail (currently just the DevicePath for each camera), use the "/devlistdetail" option.
CommandCam /devlistdetail
More than one command line option can be specified at once. For example, to capture an image from the second video capture device to a file called "output.bmp" after a 10 second delay, the following command would be used:
CommandCam /filename output.bmp /delay 10000 /devnum 2
To suppress the text normally printed to the console by CommandCam (welcome message and other information), use the "/quiet" option.
CommandCam /quiet
CommandCam error codes
If CommandCam exits due to an error, it returns one of the following error code values:
- “Error: no filename specified”
- “Error: invalid delay specified”
- “Error: invalid device number”
- “Error: invalid device name”
- “Error: invalid device serial number”
- “Unrecognised command line argument”
- “Could not initialise COM”
- “Could not create filter graph”
- “Could not create capture graph builder”
- “Could not attach capture graph builder to graph”
- “Could not create system device enumerator”
- “No video devices found”
- “No devices found”
- “Video capture device not found”
- “Error getting device name and DevicePath”
- “Could not create capture filter”
- “Could not add capture filter to graph”
- “Could not create Sample Grabber filter”
- “Could not get ISampleGrabber interface to sample grabber filter”
- “Could not enable sample buffering in the sample grabber”
- “Could not set media type in sample grabber”
- “Could not add Sample Grabber to filter graph”
- “Could not create Null Renderer filter”
- “Could not add Null Renderer to filter graph”
- “Could not render capture video stream”
- “Could not render preview video stream”
- “Could not get media control interface”
- “Could not run filter graph”
- “Could not get buffer size”
- “Could not allocate data buffer for image”
- “Could not get buffer data from sample grabber”
- “Could not get media type”
- “Error opening output file”
- “Wrong media type”
Compiling CommandCam
To compile CommandCam, you will probably require the Microsoft C++ compiler, called "cl.exe", which is included with Visual C++. Personally, I compile the program from the command line (rather than through the Visual C++ IDE) using the following command:
cl CommandCam.cpp ole32.lib strmiids.lib oleaut32.lib
It should also be straightforward to compile it through the Visual C++ GUI, but it will probably be necessary to specify the following libraries in your project’s linker options:
ole32.lib, strmiids.lib, oleaut32.lib
Source code
The complete source code for CommandCam resides in a single file. Here it is in its entirety:
// // CommandCam - A command line image grabber // Copyright (C) 2012-2013 Ted Burke // // This program is free software: you can redistribute it and/or modify // it under the terms of the GNU General Public License as published by // the Free Software Foundation, either version 3 of the License, or // (at your option) any later version. // // This program is distributed in the hope that it will be useful, // but WITHOUT ANY WARRANTY; without even the implied warranty of // MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the // GNU General Public License for more details. // // You should have received a copy of the GNU General Public License // along with this program (see the file "COPYING"). // If not, see <http://www.gnu.org/licenses/>. // // Website: https://batchloaf.wordpress.com // // To compile using the MSVC++ compiler: // // cl CommandCam.cpp ole32.lib strmiids.lib oleaut32.lib // // Last modified 24-1-2013 // // DirectShow header file #include <dshow.h> // This is a workaround for the missing header // file qedit.h which seems to be absent from the // Windows SDK versions 7.0 and 7.1. // To use the items defined in this dll, the // DexterLib namespace must be specified. // The items in question are: // // DexterLib::_AMMediaType // DexterLib::ISampleGrabber // DexterLib::IID_ISampleGrabber // #import "qedit.dll" raw_interfaces_only named_guids // For some reason, these are not included in the // DirectShow headers. However, they are exported // by strmiids.lib, so I'm just declaring them // here as extern. EXTERN_C const CLSID CLSID_NullRenderer; EXTERN_C const CLSID CLSID_SampleGrabber; // DirectShow objects HRESULT hr; ICreateDevEnum *pDevEnum = NULL; IEnumMoniker *pEnum = NULL; IMoniker *pMoniker = NULL; IPropertyBag *pPropBag = NULL; IGraphBuilder *pGraph = NULL; ICaptureGraphBuilder2 *pBuilder = NULL; IBaseFilter *pCap = NULL; IBaseFilter *pSampleGrabberFilter = NULL; DexterLib::ISampleGrabber *pSampleGrabber = NULL; IBaseFilter *pNullRenderer = NULL; IMediaControl *pMediaControl = NULL; char *pBuffer = NULL; void exit_message(const char* error_message, int error) { // Print an error message fprintf(stderr, error_message); fprintf(stderr, "\n"); // Clean up DirectShow / COM stuff if (pBuffer != NULL) delete[] pBuffer; if (pMediaControl != NULL) pMediaControl->Release(); if (pNullRenderer != NULL) pNullRenderer->Release(); if (pSampleGrabber != NULL) pSampleGrabber->Release(); if (pSampleGrabberFilter != NULL) pSampleGrabberFilter->Release(); if (pCap != NULL) pCap->Release(); if (pBuilder != NULL) pBuilder->Release(); if (pGraph != NULL) pGraph->Release(); if (pPropBag != NULL) pPropBag->Release(); if (pMoniker != NULL) pMoniker->Release(); if (pEnum != NULL) pEnum->Release(); if (pDevEnum != NULL) pDevEnum->Release(); CoUninitialize(); // Exit the program exit(error); } int main(int argc, char **argv) { // Capture settings int quiet = 0; int snapshot_delay = 2000; int show_preview_window = 0; int list_devices = 0; int list_devices_with_detail = 0; int device_number = 1; char device_name[255]; char device_serial[255]; char filename[255]; // Other variables char char_buffer[100]; // Default device name and output filename strcpy(device_name, ""); strcpy(filename, "image.bmp"); // First check if output messages should be suppressed int n; for (n=1 ; n < argc ; n++) { // Check next command line argument if (strcmp(argv[n], "/quiet") == 0) { // Enable preview window quiet = 1; } } // Information message if (!quiet) { fprintf(stdout, "\n"); fprintf(stdout, "CommandCam Copyright (C) 2012-2013 Ted Burke\n"); fprintf(stdout, "This program comes with ABSOLUTELY NO WARRANTY;\n"); fprintf(stdout, "This is free software, and you are welcome to\n"); fprintf(stdout, "redistribute it under certain conditions;\n"); fprintf(stdout, "See the GNU General Public License v3,\n"); fprintf(stdout, "<http://www.gnu.org/licenses/gpl.txt>\n"); fprintf(stdout, "\n"); fprintf(stdout, "https://batchloaf.wordpress.com/CommandCam\n"); fprintf(stdout, "This version 24-1-2013\n"); fprintf(stdout, "\n"); } // Parse command line arguments. Available options: // // /delay DELAY_IN_MILLISECONDS // /filename OUTPUT_FILENAME // /devnum DEVICE_NUMBER // /devname DEVICE_NAME // /devserial DEVICE_SERIAL_NUMBER // /preview // /devlist // n = 1; while (n < argc) { // Process next command line argument if (strcmp(argv[n], "/quiet") == 0) { // This command line argument has already been // processed above, so just ignore it now. } else if (strcmp(argv[n], "/preview") == 0) { // Enable preview window show_preview_window = 1; } else if (strcmp(argv[n], "/devlist") == 0) { // Set flag to list devices rather than capture image list_devices = 1; } else if (strcmp(argv[n], "/devlistdetail") == 0) { // Set flag to list devices rather than capture image list_devices = 1; list_devices_with_detail = 1; } else if (strcmp(argv[n], "/filename") == 0) { // Set output filename to specified string if (++n < argc) { // Copy provided string into char buffer strcpy(char_buffer, argv[n]); // Trim inverted commas if present and copy // provided string into filename char array if (char_buffer[0] == '"') { strncat(filename, char_buffer, strlen(char_buffer)-2); } else { strcpy(filename, char_buffer); } } else exit_message("Error: no filename specified", 1); } else if (strcmp(argv[n], "/delay") == 0) { // Set snapshot delay to specified value if (++n < argc) snapshot_delay = atoi(argv[n]); else exit_message("Error: invalid delay specified", 2); if (snapshot_delay <= 0) exit_message("Error: invalid delay specified", 2); } else if (strcmp(argv[n], "/devnum") == 0) { // Set device number to specified value if (++n < argc) device_number = atoi(argv[n]); else exit_message("Error: invalid device number", 3); if (device_number <= 0) exit_message("Error: invalid device number", 3); } else if (strcmp(argv[n], "/devname") == 0) { // Set device number to specified value if (++n < argc) { // Copy device name into char buffer strcpy(char_buffer, argv[n]); // Trim inverted commas if present and copy // provided string into device_name if (char_buffer[0] == '"') { strncat(device_name, char_buffer, strlen(char_buffer)-2); } else { strcpy(device_name, char_buffer); } // Remember to choose by name rather than number device_number = 0; } else exit_message("Error: invalid device name", 4); } else if (strcmp(argv[n], "/devserial") == 0) { // Set device serial number to specified string if (++n < argc) { // Copy device name into char buffer strcpy(char_buffer, argv[n]); // Trim inverted commas if present and copy // provided string into device_serial if (char_buffer[0] == '"') { strncat(device_serial, char_buffer, strlen(char_buffer)-2); } else { strcpy(device_serial, char_buffer); } // Remember to choose by serial number rather than number device_number = 0; } else exit_message("Error: invalid device serial number", 5); } else { // Unknown command line argument fprintf(stderr, "Unrecognised option: %s\n", argv[n]); exit_message("", 6); } // Increment command line argument counter n++; } // Intialise COM hr = CoInitializeEx(NULL, COINIT_MULTITHREADED); if (hr != S_OK) exit_message("Could not initialise COM", 7); // Create filter graph hr = CoCreateInstance(CLSID_FilterGraph, NULL, CLSCTX_INPROC_SERVER, IID_IGraphBuilder, (void**)&pGraph); if (hr != S_OK) exit_message("Could not create filter graph", 8); // Create capture graph builder. hr = CoCreateInstance(CLSID_CaptureGraphBuilder2, NULL, CLSCTX_INPROC_SERVER, IID_ICaptureGraphBuilder2, (void **)&pBuilder); if (hr != S_OK) exit_message("Could not create capture graph builder", 9); // Attach capture graph builder to graph hr = pBuilder->SetFiltergraph(pGraph); if (hr != S_OK) exit_message("Could not attach capture graph builder to graph", 10); // Create system device enumerator hr = CoCreateInstance(CLSID_SystemDeviceEnum, NULL, CLSCTX_INPROC_SERVER, IID_PPV_ARGS(&pDevEnum)); if (hr != S_OK) exit_message("Could not create system device enumerator", 11); // Video input device enumerator hr = pDevEnum->CreateClassEnumerator( CLSID_VideoInputDeviceCategory, &pEnum, 0); if (hr != S_OK) exit_message("No video devices found", 12); // If the user has included the "/list" command line // argument, just list available devices, then exit. if (list_devices != 0) { if (!quiet) fprintf(stdout, "Available capture devices:\n"); n = 0; while(1) { // Find next device hr = pEnum->Next(1, &pMoniker, NULL); if (hr == S_OK) { // Increment device counter n++; // Get device name hr = pMoniker->BindToStorage(0, 0, IID_PPV_ARGS(&pPropBag)); VARIANT var; // Retrieve and print device name if (list_devices_with_detail) if (!quiet) fprintf(stdout, "Capture device %d:\n", n); VariantInit(&var); hr = pPropBag->Read(L"FriendlyName", &var, 0); if (!quiet) fprintf(stdout, " Device name: %ls\n", var.bstrVal); VariantClear(&var); // Retrieve and print device path if (list_devices_with_detail) { VariantInit(&var); hr = pPropBag->Read(L"DevicePath", &var, 0); if (!quiet) fprintf(stdout, " Device path: %ls\n\n", var.bstrVal); VariantClear(&var); } } else { // Finished listing device, so exit program if (n == 0) exit_message("No devices found", 13); else exit_message("", 0); } } } // Get moniker for specified video input device, // or for the first device if no device number // was specified. VARIANT var; n = 0; while(1) { // Access next device hr = pEnum->Next(1, &pMoniker, NULL); if (hr == S_OK) { n++; // increment device count } else { if (device_number == 0) { fprintf(stderr, "Video capture device %s not found\n", device_name); } else { fprintf(stderr, "Video capture device %d not found\n", device_number); } exit_message("", 14); } // If device was specified by name or serial number... if (device_number == 0) { // Get video input device name hr = pMoniker->BindToStorage(0, 0, IID_PPV_ARGS(&pPropBag)); if (hr == S_OK) { // Get current device name VariantInit(&var); hr = pPropBag->Read(L"FriendlyName", &var, 0); // Convert to a normal C string, i.e. char* sprintf(char_buffer, "%ls", var.bstrVal); VariantClear(&var); // Exit loop if current device name matched devname if (strcmp(device_name, char_buffer) == 0) break; // Get current device path VariantInit(&var); hr = pPropBag->Read(L"DevicePath", &var, 0); // Convert to a normal C string, i.e. char* sprintf(char_buffer, "%ls", var.bstrVal); VariantClear(&var); pPropBag->Release(); pPropBag = NULL; // Exit loop if specified serial number appears in DevicePath if (strlen(device_serial) && strstr(char_buffer, device_serial)) break; } else { exit_message("Error getting device name and DevicePath", 15); } } else if (n >= device_number) break; } // Get video input device name hr = pMoniker->BindToStorage(0, 0, IID_PPV_ARGS(&pPropBag)); VariantInit(&var); hr = pPropBag->Read(L"FriendlyName", &var, 0); if (!quiet) fprintf(stdout, "Capture device: %ls\n", var.bstrVal); VariantClear(&var); // Create capture filter and add to graph hr = pMoniker->BindToObject(0, 0, IID_IBaseFilter, (void**)&pCap); if (hr != S_OK) exit_message("Could not create capture filter", 16); // Add capture filter to graph hr = pGraph->AddFilter(pCap, L"Capture Filter"); if (hr != S_OK) exit_message("Could not add capture filter to graph", 17); // Create sample grabber filter hr = CoCreateInstance(CLSID_SampleGrabber, NULL, CLSCTX_INPROC_SERVER, IID_IBaseFilter, (void**)&pSampleGrabberFilter); if (hr != S_OK) exit_message("Could not create Sample Grabber filter", 18); // Query the ISampleGrabber interface of the sample grabber filter hr = pSampleGrabberFilter->QueryInterface( DexterLib::IID_ISampleGrabber, (void**)&pSampleGrabber); if (hr != S_OK) exit_message("Could not get ISampleGrabber interface to sample grabber filter", 19); // Enable sample buffering in the sample grabber filter hr = pSampleGrabber->SetBufferSamples(TRUE); if (hr != S_OK) exit_message("Could not enable sample buffering in the sample grabber", 20); // Set media type in sample grabber filter AM_MEDIA_TYPE mt; ZeroMemory(&mt, sizeof(AM_MEDIA_TYPE)); mt.majortype = MEDIATYPE_Video; mt.subtype = MEDIASUBTYPE_RGB24; hr = pSampleGrabber->SetMediaType((DexterLib::_AMMediaType *)&mt); if (hr != S_OK) exit_message("Could not set media type in sample grabber", 21); // Add sample grabber filter to filter graph hr = pGraph->AddFilter(pSampleGrabberFilter, L"SampleGrab"); if (hr != S_OK) exit_message("Could not add Sample Grabber to filter graph", 22); // Create Null Renderer filter hr = CoCreateInstance(CLSID_NullRenderer, NULL, CLSCTX_INPROC_SERVER, IID_IBaseFilter, (void**)&pNullRenderer); if (hr != S_OK) exit_message("Could not create Null Renderer filter", 23); // Add Null Renderer filter to filter graph hr = pGraph->AddFilter(pNullRenderer, L"NullRender"); if (hr != S_OK) exit_message("Could not add Null Renderer to filter graph", 24); // Connect up the filter graph's capture stream hr = pBuilder->RenderStream( &PIN_CATEGORY_CAPTURE, &MEDIATYPE_Video, pCap, pSampleGrabberFilter, pNullRenderer); if (hr != S_OK) exit_message("Could not render capture video stream", 25); // Connect up the filter graph's preview stream if (show_preview_window > 0) { hr = pBuilder->RenderStream( &PIN_CATEGORY_PREVIEW, &MEDIATYPE_Video, pCap, NULL, NULL); if (hr != S_OK && hr != VFW_S_NOPREVIEWPIN) exit_message("Could not render preview video stream", 26); } // Get media control interfaces to graph builder object hr = pGraph->QueryInterface(IID_IMediaControl, (void**)&pMediaControl); if (hr != S_OK) exit_message("Could not get media control interface", 27); // Run graph while(1) { hr = pMediaControl->Run(); // Hopefully, the return value was S_OK or S_FALSE if (hr == S_OK) break; // graph is now running if (hr == S_FALSE) continue; // graph still preparing to run // If the Run function returned something else, // there must be a problem fprintf(stderr, "Error: %u\n", hr); exit_message("Could not run filter graph", 28); } // Wait for specified time delay (if any) Sleep(snapshot_delay); // Grab a sample // First, find the required buffer size long buffer_size = 0; while(1) { // Passing in a NULL pointer signals that we're just checking // the required buffer size; not looking for actual data yet. hr = pSampleGrabber->GetCurrentBuffer(&buffer_size, NULL); // Keep trying until buffer_size is set to non-zero value. if (hr == S_OK && buffer_size != 0) break; // If the return value isn't S_OK or VFW_E_WRONG_STATE // then something has gone wrong. VFW_E_WRONG_STATE just // means that the filter graph is still starting up and // no data has arrived yet in the sample grabber filter. if (hr != S_OK && hr != VFW_E_WRONG_STATE) exit_message("Could not get buffer size", 29); } // Stop the graph pMediaControl->Stop(); // Allocate buffer for image pBuffer = new char[buffer_size]; if (!pBuffer) exit_message("Could not allocate data buffer for image", 30); // Retrieve image data from sample grabber buffer hr = pSampleGrabber->GetCurrentBuffer( &buffer_size, (long*)pBuffer); if (hr != S_OK) exit_message("Could not get buffer data from sample grabber", 31); // Get the media type from the sample grabber filter hr = pSampleGrabber->GetConnectedMediaType( (DexterLib::_AMMediaType *)&mt); if (hr != S_OK) exit_message("Could not get media type", 32); // Retrieve format information VIDEOINFOHEADER *pVih = NULL; if ((mt.formattype == FORMAT_VideoInfo) && (mt.cbFormat >= sizeof(VIDEOINFOHEADER)) && (mt.pbFormat != NULL)) { // Get video info header structure from media type pVih = (VIDEOINFOHEADER*)mt.pbFormat; // Print the resolution of the captured image if (!quiet) fprintf(stdout, "Capture resolution: %dx%d\n", pVih->bmiHeader.biWidth, pVih->bmiHeader.biHeight); // Create bitmap structure long cbBitmapInfoSize = mt.cbFormat - SIZE_PREHEADER; BITMAPFILEHEADER bfh; ZeroMemory(&bfh, sizeof(bfh)); bfh.bfType = 'MB'; // Little-endian for "BM". bfh.bfSize = sizeof(bfh) + buffer_size + cbBitmapInfoSize; bfh.bfOffBits = sizeof(BITMAPFILEHEADER) + cbBitmapInfoSize; // Open output file HANDLE hf = CreateFile(filename, GENERIC_WRITE, FILE_SHARE_WRITE, NULL, CREATE_ALWAYS, 0, NULL); if (hf == INVALID_HANDLE_VALUE) exit_message("Error opening output file", 33); // Write the file header. DWORD dwWritten = 0; WriteFile(hf, &bfh, sizeof(bfh), &dwWritten, NULL); WriteFile(hf, HEADER(pVih), cbBitmapInfoSize, &dwWritten, NULL); // Write pixel data to file WriteFile(hf, pBuffer, buffer_size, &dwWritten, NULL); CloseHandle(hf); } else { exit_message("Wrong media type", 34); } // Free the format block if (mt.cbFormat != 0) { CoTaskMemFree((PVOID)mt.pbFormat); mt.cbFormat = 0; mt.pbFormat = NULL; } if (mt.pUnk != NULL) { // pUnk should not be used. mt.pUnk->Release(); mt.pUnk = NULL; } // Clean up and exit if (!quiet) fprintf(stdout, "Captured image to %s", filename); exit_message("", 0); }
Ted,
Thanks for the response to my original reply and your suggestions. Thanks also for the updated code. I didn’t realise that you had posted a reply as the notification of followup comments must have had a problem so I apologise for not replying till now.
I don’t know if the delay would have been necessary before or not. Certainly with the latest code it is unnecessary for me as I can simply run commandcam with no parameters at all and it works perfectly!
Thanks very much – this is much simpler than using vlc to do a still image capture as I was doing previously.
Stephen
Stephen, this program saved my life. I was using Robot Eyez with a Surface Pro 3. When I installed it a Surface Pro 4, no dice. Commandcam however works fine. Any chance of adding that very handy width and height command line from RobotEyez? Keep up the good work
Hi Jeff,
Unfortunately, I ran into problems when I tried to add an option to CommandCam to set the image resolution. As I recall, that was actually on of the reasons I began re-writing the whole thing using a different structure (the re-write became RobotEyez).
I believe it may be possible with some cameras to adjust the default capture resolution for the device using a different tool (e.g. one supplied by the camera manufacturer) and then CommandCam will capture at that resolution. I suppose that’s worth investigating with your camera. Alternatively, you could use a command line tool like ImageMagick to automatically change the resolution after the picture has been taken. Obviously that won’t produce a great result if you scale it up a lot though.
Ted
Hi Stephen,
Thanks for getting back to me. I’m delighted to hear the updated code is working well for you. Actually, your original feedback prompted me to clean up several bits and pieces in the program, which has produced a much more satisfactory version, so thanks for that.
Ted
Interesting product.
May I ask: I would like to automate testing, is there a way to start and stop (and filesave) a video from the webcam? (not just an image)
Hi Bill,
Unfortunately, CommandCam does not record videos. However, I have been considering writing a little program to do just what you describe (it’s certainly possible to do it and I can probably reuse some of the CommandCam code). I might get a chance to start writing it over the weekend. If you can describe a little more about the way you would be using it (how many videos, recording how often, how many cameras, etc) I can try to keep that in mind when I look at it.
Regards,
Ted
hello, thanks for your new version of snapz. it works now. the only thing i wish i could influence would be the size of the captured image. my cam can go up to 1920×1080. an additional switch to choose some image sizes would be perfect.
happy xmas,
sheldon.
Hi Sheldon,
I was originally hoping to include this feature (image resolution selection), but it turned out to be quite tricky to make CommandCam do it due to the way it interacts with DirectShow (the underlying video API provided by Microsoft). However, I have since written another little program called RobotEyez which has similar functionality to CommandCam, but interacts with DirectShow in a way that makes it slightly easier to specify the image resolution. Perhaps I could compile you a version of RobotEyez that selects the specific resolution you want to use? Would that help if I could provide a version that does one specific resolution of your choice, or are you just interested in being able to control the resolution in general (e.g. 1920×1080)?
Ted
yes Ted do that 🙂
Great little tool! The only thing that would make it better is incrementing file names. I have it running in a batch file triggered by a motion detector. The motion detector is rather quick, so the pic gets written over before I get to see it. I only get to see the last pic. Other than that GREAT TOOL!!!!!!
Hi Robert,
Thanks very much for the positive feedback – it’s much appreciated. I’ll keep your suggestion about incrementing filenames in mind and try to add that feature the next time I get to spend a bit of time on CommandCam. In the meantime, you should be able to create the same effect by “wrapping” CommandCam in a batchfile that checks what filenames are already present before running CommandCam with an available filename. I’ll have a bit of a think about it and try to post an example later.
Regards,
Ted
You can create a for loop in the bat script that adds to the file name. I’ve got one that triggers every half hour on scheduelled task and saves 4 images.
The double % or %% is to make it literal.
“FOR %%i IN (01, 02, 03, 04) DO C:\temp\CommandCam.exe /delay 30000 /filename C:temp\Cam02_%%i.bmp”
I’m sure this can be modified to add a timestamp
Thanks Kerber – that’s a handy tip!
Ted
Ted,
Using the Set command I added a line that takes out the unwanted junk from the %date% and %time%. and added it to the batch file as %DNT% . then i use the /filename %DNT%. Now it works like a charm! Thanks!
Brilliant! Thanks very much for the useful tip. I’m sure it will help other users!
This worked beautifully to silently capture an image of the kid that stole one of work computers. Instead of doing a remote screen takeover which would have alerted the user, capturing an image behind the scenes worked great. Excellent evidence and the computer was recovered by police!
Bravo! I hope you got the computer back. Happy Christmas!
Regards,
Ted
Great program, would be awesome if you could add jpeg support in the future as well!
Thanks Paul, I’ll give it a shot.
Good work!
Would it be possible to wait in the preview mode until the user presses a button or clicks in the preview window.
Hi Nils,
Thanks for your comment. I’m not sure how straightforward it will be to add this feature, but I’ll add it to the list and see if it’s possible.
Regards,
Ted
Used this one in an hta + vbscript app. and it works great xcept I can’t get a preview before snapping up the picture. As for jpeg conversion, i used the irfanviewportable. it is kinda lighter-weight compared to imagemagick and can also be run thru commandline,
Hi John,
Thanks for your message. I’m glad to hear you found CommandCam useful. Regarding the preview window not working, you could try another of my programs called RobotEyez which has many similar features, but performs the preview and capture in a slightly different way behind the scenes. You might find that it works better for you. Here’s the link:
https://batchloaf.wordpress.com/2011/11/27/ultra-simple-machine-vision-in-c-with-roboteyes/
I’ll probably merge CommandCam and RobotEyez into one program in the near future, since I’m really just using RobotEyez to experiment with new features.
Thanks,
Ted
KOOL!
there is web site sensr.net
allows to post pictures from wifi web cameras
but i have regular USB cam
so using your CommandCam.exe i can emulate “wifi web cam”
kind of “capture img, upload it to sensr.net”
Great, I’m glad you found it useful! That website is funny – they seem to be all about keeping an eye on your pets over the internet. I never thought of that before.
Wow, this is exactly what I’ve been looking for. THANK YOU.
I’m using Logitech Quickcam pro 9000 camera. I was wondering it’s it’s somehow possible to go beyond 640 x 480 resolution? It’s possible with their bundled software but I guess it can’t be done with command line?
Does anyone know which web cam gives the highest resolution images natively with CommandCam?
Thanks! 🙂
Hi Jani,
Thanks for your comment! You’re not the only one looking to use CommandCam at higher resolution and there’s no major technical problem to making the change to the code – I’ve just been flat out at work, so I haven’t had a chance to do it. Hopefully, I’ll add the feature very soon.
By the way, is there a particular resolution you want to use it at?
Regards,
Ted
Hi Ted
Thanks for being so helpful! I just want as many pixels as possible. I’m working on a panorama project where I combine two webcam inputs into one panoramic one. Due to lens correction, cropping & other functions I perform on the images, I lose some pixels – and there’s nothing I can do about it. So, the more I have in the beginning the better. I was wondering if anyone knows what camera gives the highest resolution with CommandCam. If there’s no solution to go higher, that’s fine if I know what to buy to get more pixels. Many cameras claim 1920 x 1080 but I don’t want to spend the money if I get back 640 x 480 LOL! 🙂
Thanks!
Well, what I’m planning to do is add a command line option for requesting a specific width and height for the captured image and it will just depend on the camera (and its driver) whether that resolution is supported. Hopefully, that will solve your problem.
By the way, one feature that another user requested is the ability to capture from multiple cameras simultaneously, so I’m going to have a go at supporting that. Perhaps that might be useful for your panorama idea too.
Regards,
Ted
HI – great program. Just some input:
I recently bought a logitech C270 web camera (<$20).
Although it claims to be widescreen with 720p support I have done quite a few trials and the best/most detailed picture quality is when using it as 4:3 ratio at "3mp". This translates to 2048×1536.
Interestingly:
1) Choosing 640×480 results in a sharp picture but other supported resolutions between those two listed above are actually worse quality….
2) Choosing widescreen modes results in a cropped image compared to a 4:3 pictures so the sensor is probably 4:3 nativley.
So I would very much appreciate support for the above resolution (2048×1536) and I think it may be beneficial to owners of a lot of similar webcam models*
* There appear to be a lot of similar logitech models with differing feature sets but the housing is very similar so there is a good chance the sensor is the same in all of them.
I will be using command cam at home for a security system and also (potentially) at work from some remote monitoring of my experiments. I just made a little donation – thank you.
Hi W3lshboy,
Thanks so much for your comment. Sincere thanks also for donating – if only more people would 😉 I’ll do my best to add that feature to facilitate capture at the resolution you mentioned. Watch this space!
Thanks again,
Ted
I finally got around to trying to integrate the latest RobotEyez today with my C270 logitech web cam. The max resolution I seem to be able to get working for RobotEyez capture is 800×600. From the logitech app I can capture at 3MP and the windows properties for that JPEG is 2048×1536. I have also tried RobotEyez at 1024×768. Both (high) resolutions report “could not render video preview stream”. interestingly ffdshow pops up a dialog when i try to capture but both “use ffdshow” and “do not use ffdshow” result in failure. I also tried a 4000ms pause for the higher res grabs. (PC: Windows 7 64bit.)
Aside: Perhaps this API (“IAMStreamConfig::GetStreamCaps”) could be integrated into RobotEyez to assist with this issue and debug.
It would be good to know what video resolutions are available on a per webcam basis:
http://msdn.microsoft.com/en-us/library/ms787898(v=vs.85).aspx
Hello again W3lshboy,
Thanks for the update. Here’s my (so far unspectacular) update:
I’ve been trying a different approach for capturing higher resolution still images from USB cameras such as yours, using the Windows Image Acquisition (WIA) API. I’ve made some progress, but I haven’t yet managed to actually grab an image to a file (or at all – I’m not quite sure yet). I’ll post my current code in a new blog post in a minute so that you can take a look if you’re interested. Hopefully, I’ll get this working soon and it will provide a better way to capture still images at the maximum resolution supported by each camera.
Ted
Hi batchloaf
I was wondering where would i find the libraries for this project.
Which libraries do you need? I have Microsoft Visual C++ 2010 Express installed. Assuming it’s still available, it’s a free download. If you’re getting an error when you try to run it, perhaps you could post the exact error here?
I was talking about these .lib files:
ole32.lib, strmiids.lib, oleaut32.lib
Below are the linker errors that i am getting:
1>—— Build started: Project: camera, Configuration: Debug Win32 ——
1>Build started 2/23/2012 2:44:32 PM.
1>InitializeBuildStatus:
1> Touching “Debug\camera.unsuccessfulbuild”.
1>ClCompile:
1> All outputs are up-to-date.
1> All outputs are up-to-date.
1>ManifestResourceCompile:
1> All outputs are up-to-date.
1>camera.obj : error LNK2001: unresolved external symbol _FORMAT_VideoInfo
1>camera.obj : error LNK2001: unresolved external symbol _IID_IMediaControl
1>camera.obj : error LNK2001: unresolved external symbol _PIN_CATEGORY_PREVIEW
1>camera.obj : error LNK2001: unresolved external symbol _PIN_CATEGORY_CAPTURE
1>camera.obj : error LNK2001: unresolved external symbol _MEDIASUBTYPE_RGB24
1>camera.obj : error LNK2001: unresolved external symbol _MEDIATYPE_Video
1>camera.obj : error LNK2001: unresolved external symbol _CLSID_VideoInputDeviceCategory
1>camera.obj : error LNK2001: unresolved external symbol _CLSID_SystemDeviceEnum
1>camera.obj : error LNK2001: unresolved external symbol _CLSID_CaptureGraphBuilder2
1>camera.obj : error LNK2001: unresolved external symbol _IID_ICaptureGraphBuilder2
1>camera.obj : error LNK2001: unresolved external symbol _CLSID_FilterGraph
1>C:\Users\drmaq\Dropbox\CAMERA\camera\Debug\camera.exe : fatal error LNK1120: 11 unresolved externals
1>
1>Build FAILED.
1>
1>Time Elapsed 00:00:03.68
========== Build: 0 succeeded, 1 failed, 0 up-to-date, 0 skipped ==========
On my PC, strmiids.lib, ole32.lib and oleaut32.lib are all in the following location:
C:\Program Files\Microsoft SDKs\Windows\v7.1\Lib
So, they’re part of the Windows SDK. As far as I recall, the Windows SDK is installed automatically with Visual C++ 2010 Express, but if not it can be downloaded from MSDN.
It worked thanks
Great, I’m glad you got it working.
OK i have developed another problem
it seem to build fine but when it executes it only displays this error:
“Error opening output file”
This is the output form the visual C++ debugger:
‘camera.exe’: Loaded ‘C:\Users\drmaq\Dropbox\CAMERA\camera\Release\camera.exe’, Symbols loaded.
etc etc
Is it possible that you have a previous output file of the same name open in another program? For example, if you run CommandCam once and it produces an output image file which you open in an image viewer program, CommandCam won’t be able to create an output file of the same name as long as you have the first output file open. Could that be why you’re getting this error? Do you have an image file open in another window perhaps?
Nope i deleted the image.bmp file and move the exe file to a new folder and the i still get the output file error
thanks I will look in to that
but the exe i got from your site works perfectly
Hmmm, I’m not sure what’s causing the problem. If your libraries are in the SysWOW64 folder, then I’m guessing you’re running Windows 7. I’m still using XP, so perhaps your exe has compiled somewhat differently to how mine did. Sorry I can’t suggest anything more helpful, but since I don’t have Windows 7 here, it probably isn’t possible for me to reproduce your problem and find a solution.
Anyone else have any ideas?? If the exe you downloaded is working ok, my guess is that it must be a compile settings issue.
Hi Ted,
I have been searching for webcam image capturing app for my little project (for home server). Finally I found your app. which is simple and cool at the same time.
I have written a little application which can start your app., wait for the image, and finally convert it to a jpg file. (Convert is necessary because of my slow upload speed.)
The result is: http://92.249.157.109:1080/take.html
If I had time I would merge the two codes (yours and mine and add some switches e.g. jpeg, png).
Or I can send you my little program c++ (with gdi+) source if you want to use it.
Thak you for your app.
Best Regards,
Nandi
very good program, but I have one problem or one question. If I run skype video conference or run any other program which is using a webcam in such time when I try to run CommandCam.exe, I get error (Error: 2147943850, Could not run filter graph). So my question: it is possible to make a webcam shot while a webcam is active with other program sush as skype and etc.. ? sorry for my poor english language, I hope you understand what I want to ask.
Hi Bem,
I’m afraid CommandCam cannot open a camera that is already being used by another program because Windows will not allow it. If the camera picture is displaying in the Skype window, perhaps you can do a screenshot instead? Did you try pressing Alt + Prt Sc to see if the image can be captured from the screen? If it can, then there’s probably some way to trigger the same screen capture process from a command rather than a physical key press.
Ok. other question. It is possible to turn off a webcam led light while a webcam is active ?
It’s not possible with CommandCam. Individual camera drivers may provide some way of doing it, but I would imagine that most force the light to be turned on while the camera is active.
Mr. Burke, the Logitech C920 boasts 15MP resolution via some sort of post processing of the data gathered. Questions: 1. Does commandcam.exe support the C920? 2. Is it possible to integrate the post processing of the data into a hi-res 15MP BMP file in a batch mode first calling commandcam then the post processor? 3. Is it possible to integrate the post processing into the commandcam.exe?
Oh… one more question. Using your software what resolution could I expect without any post processing of the data from the C920?
Hi Ken,
Ok, I’ll do my best to answer your questions:
1. CommandCam accesses the camera via the DirectShow API, so any camera with a DirectShow driver should work (that includes most USB cameras). I’m not sure about the C920 specifically, but CommandCam is free so all you have to do is give it a try!
2. I’m afraid I don’t know anything about the post-processing for the C920. How is it done normally? If they provide a command line tool for doing it, then yes, you should be able to create a batch file that calls CommandCam first followed by the post-processor. However, if the post-processing is encapsulated within the camera driver and simply used when the camera is opened for high resolution capture, then CommandCam won’t currently let you control whether or not it’s used. CommandCam currently just uses each camera’s default resolution, so unless the camera defaults to high res, you’re out of luck. I am working on a re-write which will allow you to specify the resolution. You can check out an early version of this (called RobotEyez) here.
3. I won’t be integrating any features into CommandCam that are specific to a particular camera model. If it’s integrated into the camera’s DirectShow driver, then it might be used during high res capture, but only if it happens automatically – I won’t be programming it specifically into CommandCam.
4. The capture resolution just depends on the camera and th driver. As far as I remember, nothing in the code of the current release of CommandCam specifies the capture resolution, so the default resolution is simply used. On my camera, that’s 640×480, but I don’t know about other cameras. As I mentioned before, just give it a try with your camera and see what resolution it used.
I’ve been looking for a simple program to capture webcam images and commandcam seems to be the best thing out there. We want to use commandcam to capture images of growing plants for a science education video, but would need to be able to bump the image resolution to 1920 x 1080 (i.e. HD quality). Do you have definite plans to add that functionality?
Hi Michael,
What a fascinating application! Since it’s for an educational cause, I couldn’t resist leaping into action and trying to add this feature. I’ve been slowly working on a rewrite of CommandCam, which for some reason I’m calling RobotEyez until it’s ready to replace the original. Anyway, it’s basically working and I’ve just added command line flags to set width and height. Please take a look at my blog post about RobotEyez. You can download the current RobotEyes executable from the link on that page.
Basically, to do what you want, you’ll probably just run it as follows:
RobotEyez /width 1920 /height 1080 /bmp
With some cameras, it seems to help to leave a short delay for warming up before snapping the photo. For example, this will add a 5 second delay:
RobotEyez /delay 5000 /width 1920 /height 1080 /bmp
RobotEyez has a bunch of other command line flags that are documented (very briefly) on the blog post I linked to above.
Ok, I hope that helps. Please let me know if it works for you or not. I haven’t been able to test it up to 1920×1080 since my camera doesn’t support that resolution, so I’d love to know if it works. Also, if you do manage to get any videos of the plants growing, I’d be interested to see them!
Regards,
Ted
Ted-
Thanks so much- we are implementing the software now and hope to have videos of plant development for show students in a few months.
M
Bravo! I hope it works out well for you.
Mr. Burke, my Logitech C920 arrived today. It really does provide ~15mp resolution. Unfortunately, the post processing or interleave processing of multiple samples happens within one GUI application. You must have the GUI panel up, and you must push the button on the GUI panel. The finest in “graphical computing”. I have tried both your sets of tools and everything available on the web. Best can do without Logitech’s software is 680×480.
I’m looking at scripting HDR-type images as the dynamic range of the camera itself if rather weak. Is there any way to specify either manual exposure values, or better yet, an exposure offset? (read in the measured exposure and adjust from there) – bracketed exposures.
Hi Eric. CommandCam does not provide any way to set the parameters you mention. To be honest, I don’t even know whether it’s possible to configure those parameters via DirectShow (the Microsoft API that CommandCam uses to communicate with the camera). However, I doubt it would be straightforward to introduce this functionality into CommandCam, so I currently have no plans to attempt it. Also, since CommandCam opens the camera device for video capture (even though it’s just snapping a single frame), it’s probably not ideally placed to control this kind of thing. Sorry, that’s probably not much use to you.
How to check with c++, if the webcam is active now or not ?
Hi Bem. I’m not sure what the best way to do this it, but it may not be all that easy to check. My suggestion is to try opening the device (look at the CommandCam source code to see how) and see what happens. It’s possible that DirectShow provides some other way to check, but I’m afraid I don’t know for sure.
Hey, Batchloaf,
Could I use your commandcam.exe in my software? Just the commandcam.exe would me named cc.exe. That would be realy great!!! Sorry for my bad language…
Hi Titus. Thanks for your message. Can you give me some more details please? What is the software you’re writing and are you going to be selling it? Would I be credited in some way?
Hellow, batchloaf. The software I am creating will be some kind of monitoring program. A legal keylogger maby… But I am realy looking forward to make some money from it… I can tell you nothing about krediting for now… Becouse I just started to create my program… I am not sure for the future and I can’t tell you about crediting for now… But if you give me your email I will keep you up to date with my program… Thankyou… What about the permission?
Hi Titus. So, you’re writing some sort of keylogger that you hope to make money from, but you’re not planning to pay me or give me any credit. Hmmm. I hope you’ll understand why I might not be all that keen on giving you open-ended permission to use my software however you want under these circumstances 😉
I have been intending to explore the possibility of making CommandCam available under a GPL license (or similar), but I just haven’t had time to investigate it properly. If I do decide to do that, I suppose you’ll be able to bundle CommandCam with your software subject to the terms and conditions of that license.
Tool is useful. But you can mode when capture by webcame, don’t use flash .
Capture do not flash in webcam.
Do you mean that you want to take a picture from a camera that has a flash bulb attached? If so, it probably won’t be possible to control this from CommandCam because it opens the camera as a video capture device rather than for still capture.
A fantastic little utility. I intend to use it at our school when the new intake of students start in September.
I needed something where I could take a photo and rename the file and your app makes it both quick and easy so thank you very much.
Thanks Paul! I hope it works out for you. I actually originally wrote this to help with a project that one of my own students was working on. Since then I’ve used it quite a bit myself for simple machine vision exercises in a Robotics class that I teach. Let me know how you get on with it next term.
I’m using your software with Alpha Five (database), and it works great! Thanks.
OK, here’s a challenge – have you done anything like this for a signature pad? I’d be willing to pay for a solution for this problem. Please contact me at Compunique@charter.net.
Bill
Hi Bill. Do you mean the electronic tablets for capturing signatures in point of sale systems and the like? If so, I haven’t worked with these before, but I would assume that most hardware vendors already provide some sort of software to do this. I’m sure it can be done, but I’m not sure how transferable the solution would be from one brand of tablet to another.
Nice tool! So much lighter than VLC for taking webcam snapshots. I have a couple of small feature requests:
1. I found that I can control how long the preview window is present through the /delay option. Not sure if this was intended, but it’s useful for letting the user adjust focus, lighting, etc. Could the code be changed or an option added to leave the preview window up until the user closes it? While the window can stay up longer, the video stops beforehand, leaving an empty frame.
2. After a successful image capture, the following text is returned:
CommandCam Copyright (C) 2012 Ted Burke
This program comes with ABSOLUTELY NO WARRANTY;
This is free software, and you are welcome to
redistribute it under certain conditions;
See the GNU General Public License v3,
https://batchloaf.wordpress.com
This version 21-4-2012
Capture device: HP Webcam
Capture resolution: 640×480
Captured image to snapshot.bmp
Since I’ll be calling your program from within a Python script using the subprocess module, a nicer result would be nothing or 0 for successful completion and a 1 or error message if something goes wrong. So perhaps some additional options like /help, /version, /license, /verbose, etc. would accomplish the same thing?
Thanks again for sharing this with the community!
Hi Dan. Thanks for the suggestions. They seem like good ideas, so I’ll try to incorporate them when I next sit down to do a bit more work on CommandCam. Once the summer holidays finally arrive, I’ll hopefully get a chance to spend a day or two giving CommandCam a complete overhaul and incorporate these and other features.
Hi Ted,
Just wondered if you had considered these suggestions from last June.
Thanks!
Dan
Hi Dan,
I’ve added a “/quiet” option which allows normal text output from CommandCam to be silenced. Error messages will still appear however.
I’m not too sure what to do about leaving the preview window open. That’s a more significant change and I would need to be think the repercussions through carefully. Since I’m hoping to redo CommandCam when I can find the time (probably using the RobotEyez code rather than the current CommandCam code), I don’t think I’ll change this aspect of it right now.
Ted
Sounds good!
Dan
Is posible to save in jpg rather han bmp?
It is not currently possible to save directly to jpg in CommandCam. However, you can easily use ImageMagick‘s “convert” command to automatically change the bmp file into a jpg as soon as it’s saved. ImageMagick is free to download and provides an amazing set of command line tools for image conversion and modification (scaling, colour converting, adding text, etc). Once you’ve installed ImageMagick, all you need to do to convert CommandCam’s saved bmp file into jpg is something like this:
I’m trying to capture the device list to a file using redirection:
c:\apps\CommandCam.exe /devlist > c:\tmp\devices.txt
The device list is printed to the console window but the file is empty. Any hints?
Hi Dan,
Yes, Ian (below) has hit the nail on the head. Because CommandCam prints the device list to stderr rather than stdout, you need to redirect the output slightly differently – using “2>” instead of “>”. Upon reflection, the device list should probably just print to stdout, so I may change that in the future. For the time being however, Ian’s suggestion should work fine. i.e. something like the following…
Thank you both for the quick reply. With this bit of information, I was able to capture the list within my Python app for building a drop-down menu. For me, it would be cleaner to have normal outputs come to stdout and error messages to stderr along with exit codes like 0 and 1, but I can certainly live with the current behavior.
Hi Ted,
Did this ever get cleaned up? Exit codes would be nice too.
Thanks!
Dan
Hi Dan, I’ve just updated CommandCam so that all output text other than error messages goes to stdout, which can be redirected in the normal way if desired. Also, it now returns a unique error code for 34 separate errors that can cause it to exit (although I’m still testing this bit).
Thanks! I’ll watch for the next release.
Dan
Dan, try c:\apps\CommandCam.exe /devlist 2> c:\tmp\devices.txt
Useful utility. Thanks for your efforts.
I was able to switch on the preview window but is it possible that the image is not captured until the user accepts it.
I tried this app. and is working fine with the integrated webcam on the system, will it work fine with plug and play devices i.e. usb cams?
Is it possible to capture images from multiple devices on a single click?
Looking forward to your reply.
Hi Sawyer,
Thanks for the feedback. To answer your questions:
“is it possible that the image is not captured until the user accepts it?” It would be possible to do this, but I haven’t previously regarded it as a big priority since most people use CommandCam within some kind of automated system or script where there won’t be interaction with a user while it runs. There are already many other cam snapshot programs that let you take a snapshot with a mouse click on a GUI. The reason I wrote CommandCam was to facilitate taking a snapshot from a script for an automated system. I’d be interested to hear what you’re using it for – maybe you’re using it in a scenario i hadn’t thought of, in which case I might consider including it.
“will it work fine with plug and play devices i.e. usb cams?” It should work with most plug-in USB webcams. That’s what I’m using myself. CommandCam uses the DirectShow API to access the cam, which is currently the standard way on Windows and the method supported by the vast majority of USB cams.
“Is it possible to capture images from multiple devices on a single click?” You’re not the first user to ask about simultaneous capture from multiple devices. I conducted a preliminary investigation into doing this and I’m pretty sure it’s possible, but I haven’t found the time to do it yet. Once term ends (I’m a lecturer), I’m hoping to spend a couple of days rewriting CommandCam to incorporate many of the features that users have suggested. Hopefully, that will include simultaneous capture from multiple devices.
I should point out that of course it is currently possible to capture from multiple cameras one at a time. Just add several CommandCam commands to a batch file and then run the batch file. There will be a short delay (one or two seconds?) between the images captured though, so they won’t be simultaneous. Here’s what the batch file might look like:
Paste those lines into a plain text file (e.g. in Notepad), save it as something like “multicam.bat”. You’ll then be able to capture from multiple cameras (one at a time) by running “multicam.bat”.
Hi Batchloaf
Thanks for taking out time to reply.
According to the initial requirement, which was exactly the same which this tool does now.
Now the requirement is changed, user should be able to see the preview image from all the cams attached to the system on a single window split into number of cams attached and after that on a single click individual images from each cam should get stored at a particular location. The requirement for the preview of the image is because the camera might be adjusted before clicking the pic.
If you know some other tool which can be used for this please let me know.
Hi Sawyer,
I haven’t ever used a program that does exactly what you describe, although I would imagine that such programs must exist. Perhaps you could look for a CCTV / security camera monitoring application?
Anyway, I’ll keep your application in mind when I do the next phase of development on CommandCam. I can’t make any promises, since I’m just not sure how awkward it would be to preview multiple cameras that way. As a matter of interest, how many cameras are you planning on using?
Regards,
Ted
Hi Ted
The cams will be placed at a security check post. One of the cam will focus on the registration number plate of the vehicle, one on the driver of the vehicle, so at least 2 cams will be used. The registration number plate of the vehicle might be positioned differently on different vehicles that is why the user should be able to adjust the cam before capturing the image.
I’ve been looking for a program to take a webcam shot for ages, and I’ve finally stumbled across this little gem! This is fantastic. I know you’re probably overwhelmed with feature requests at this point, but I was wondering if you would be able to add a /silent switch? As it is, when the program is called, it gives an output in a command prompt window. What I’d like to use this program for is theft recovery for the school district that I’m working for and to have the program activated remotely via other programs we have on the computer. However, when the programs activate CommandCam as it is, it gives a prompt. If you could provide any feedback on the possibility of making a /silent switch, or how I might go about doing this, it would be very much appreciated! Thank you so much.
Has there been any update to this? Something like that would be pretty useful for how I’d be using it.
Hi Josef & Dylan.
Sorry, no update on this yet. The teaching holidays have just started here in DIT, so I’m finally back on research mode for a couple of months. I’m therefore optimistic that I’ll be able to squeeze in a couple of days of concentrated CommandCam development some time over the next few weeks, at which point I’ll hopefully incorporate this and other requested new features.
Ted
It works great in windows 7 64-Bit but i didn’t understand how it identified the web-cam ?? as there are many types of Web Cam ,
I have Sony VAIO Laptop and when i run your CommandCam it successfully capture the image, i couldn’t understand how it identified what kind of webcam i am using ?
I’d be glad if you can add few guidelines explaining the idea behind your program and how you start coding it ? anything that would help me understand the algorithm would be appericiated
Thank you
Hi 0x3ff3ct3d,
Sorry, I forgot to reply to this when you first posted it. The basic structure of the part of the program that opens the webcam is based on the code provided by Microsoft in the DirectShow documentation on MSDN. To be honest, the way you open cameras in DirectShow seems very over-complicated to me (actually, I find that with all COM programming), but that’s just the way it is. I’m not even going to attempt to explain any of that stuff about device enumerators, monikers, property bags, etc. I just kind of felt my way through it by following the MSDN documentation.
Here’s a link to the section of the MSDN documentation that I followed:
http://msdn.microsoft.com/en-us/library/dd407331%28v=vs.85%29
Best of luck making sense of it! (Not an easy task.)
Ted
and does it possible to code it in VBScript ?
I think you’re asking if it’s possible to use DirectShow in a VB Script program – is that right? If so, then yes, it’s probably possible since DirectShow is a COM API. However, I don’t really have any experience of VB Script programming, so I’m afraid I can’t offer any advice on it.
Here’s a link to a discussion I came across when I googled “DirectShow VBScript”:
https://groups.google.com/forum/?fromgroups=#!topic/microsoft.public.scripting.vbscript/ZzW23qXOFBI
Hi
I tried your program, on a Dell Inspiron N5050 with Integrated Webcam. Sadly it does not work, rather it is stuck and the line says
Capture Device : Integrated Webcam.
Nothing happens at this point and no image is saved.
Please tell me what I’m doing wrong.
Thanks.
Sorry for the false information.
The program now works, I believe it was a Dell Software Problem.
It’s amazing really.
I have now used it in AutohotKey (another language) to take a picture from the WebCam whenever somebody runs the laptop.
Also, I use AHK to lock the system and then CommandCam is used to take a photo of anyone who types the wrong password.
I thank you for your work.
Hi Rejjy,
Thanks for your comments. That’s great that you got it working! Best of luck with developing your security system.
Ted
I don’t suppose you have had the time to force the console window to not appear?
Hi m,
CommandCam is a console application, so it always displays a console window when you run it, in order that the output text is visible. I believe it is possible to get a console application to run in an “invisible” console window, but I haven’t tried that myself. Here’s a link outlining the general principle:
http://stackoverflow.com/questions/492876/how-to-hide-the-console-of-batch-scripts-without-losing-std-err-out-streams/493074#493074
Alternatively, you could modify the CommandCam code and recompile it as a Windows application. To do so, you would need to change the “main” function to “WinMain” and possibly include an event loop or something like that. I don’t have any plans to do this in the immediate future.
Absolutely Fantastic! Nice work!!!
Thanks João! Much appreciated.
Ive been searching for something like this for days! Thank you so so so much
Thanks Josh! I originally wrote it because one of my students needed a program like this for his final-year project (he was taking snapshots of pigs in pens to track their weight – long story). I was absolutely amazed that we couldn’t find an existing program that did this in Windows. Anyway, I’m delighted to hear that you’re finding it useful – best of luck with whatever you’re creating.
Thank you so much. I have been looking for a simple lightweight webcam capture program for a long time. Most programs just won’t work with my webcam and the few that work are resource hogs.
Hi John. You’re welcome – glad to hear you’re finding it useful!
Thank you batchloaf, great program !
Is it possible to include another option maybe a “snapshot button” to snap during preview instead of a delay ? maybe as an option ?
442 // Wait for specified time delay (if any)
443 Sleep(snapshot_delay);
Guido
Pause & commandcam.exe, Might this work?
Hi, very useful program!
Can you please add a function int getColor(int x, int y) that returns the RGB for the selected pixel?
Thanks 🙂
Hi Alessandro,
Do you mean add the C function to the program? If so, how does the function get called? Where does the pixel value go? Or do you mean that someone could run CommandCam at the command line and instead of getting an image file, three numbers (R, G and B for the specified pixel) just get printed?
Yes I mean add the C function to the program, i will compile your program as a library, i.e. under windows i’ll compile it as a DLL to be used with another program i’m developing 🙂
I think that this is not gonna be considered, am i wrong?
Hi Alessandro,
I still don’t completely understand what you want to do. If you just want to check the colour of one pixel, I’m not sure that adapting the CommandCam code is going to be the ideal way to do it – it would probably be quite a slow way to get the colour of a single pixel. However, if you’re really determined to try it, the first place I would look is around line 520 of the code listed above, just after the line “CloseHandle(hf)”. At that point in the program, the captured image has just been written to a file and the raw pixel data is still stored in the array pBuffer. To print the RGB colour of pixel (x,y) at that point in the program would be quite straightforward – you could just add something like the following lines of code (I’m using the centre pixel of a 640×480 image as an example):
If you’re compiling the CommandCam code as a DLL instead of an application, I suppose the function you’re talking about will just be the main function (renamed of course). The code above just prints the pixel value in the console, but if you want to return it from your function, you’ll have to work out some way of doing that.
Ok thank you, just another question, is there a way to get directly the HSB values instead of RGB, without converting?
On lines 383-384, I specify the colour format as 24-bit RGB (one byte per colour component) when setting the media sub type for the SampleGrabber:
You could look up MSDN to see whether a HSB sub type could be specified here instead. It’s possible that there is also a sub type for HSB. If there is and if it’s a 24-bit format, you might be in luck. DirectShow tries to deliver the colour format you request (inserting converter filters if required and available), but if you’re looking for an unusual format, I’m not sure how reliable it would be using different cameras – you might find that it worked with some but not with others. RGB24 has the advantage of being pretty standard, so that even if it’s not the format that the camera driver outputs, it should always be possible to convert to that.
Since you’re only getting one pixel, surely it’s not that big a deal to just do the conversion manually though? It might be easier than messing around with the DirectShow colour format.
Actually, I’ve just had a quick dig around on MSDN myself and I don’t see any HSB or HSV colour format there. There are plenty of RGB (and similar formats) and a selection of YUV formats – would they be of any use to you? The Y component of YUV is probably very similar (or identical) to the B component of HSB, but the colour components are different to hue and saturation.
You can see the list of supported formats here:
http://msdn.microsoft.com/en-us/library/windows/desktop/dd407353%28v=vs.85%29.aspx
The two most relevant sections are “Uncompressed RGB Video Subtypes” and “YUV Video Subtypes”.
Hi, I’m using the commnad.exe file to capture images off my HP webcap for my access app. When I run the commnad.exe on its own it captures and stores the image in the same folder as the commnad.exe file. But when I shell it from access, it doesnt. I can see the webcam light flicker but it does not seem to capture. Any help?
Hmmm, I’m not sure. When you say that you’re shelling it from access, do you mean that you’re running it from some kind of Visual Basic macro in MS Access, the database application? I don’t have Access installed on my PC, but I can try launching CommandCam from Excel to see if that works. The only thing that springs to mind that might prevent it from working correctly is that maybe the working directory is set to a location that CommandCam isn’t allowed to write to. Could that be the problem?
Thanks for the quick reply. I’m shelling in VBA. Below is my code:
Dim img
img = Shell(Application.CurrentProject.Path & “\CommandCam.exe”, vbMaximizedFocus)
When I run it on its own, the image is saved in the same location as the CommandCam.exe. My app is also located in the same folder so I expected the same behavior when I shell CommandCam. I saw other documentation on how to set certain parameters such as delay, folder location etc but I have not figured out how to do that. Maybe if I can change the save location or even preview the image then I might have the chance to save it where I want.
Thanks again for your help.
Hi Percy.
Where do you want to save the image? Do you want to save it in the same directory as your VBA application? If so, I think it’s ok to just put CommandCam.exe in the same folder and make sure the working directory is set to that location before you use shell to run it.
The way you’ve written your shell command, it looks as though you’re trying to store the captured image into the img object, but if that’s what you’re trying to do, it won’t work that way. CommandCam just saves the image as a file – it doesn’t “return” an object that you can assign to a variable/object this way. Instead, you should just run the shell command, then use another line or two of VB to load the image from the file and store it in your img object. I’m sure there’s a function called LoadImage or something like that in VB which will do this for you – I’m just guessing though, because it’s been a while since I wrote any VB.
Ted
“make sure the working directory is set to that location before you use shell to run it.” How do I do this?
Well, it depends. If the working directory is not already set to the folder you want, you’ll need to google to find out how. However, I’ve just tried running CommandCam from VBA in Excel and it turned out that I didn’t need to change the working directory at all, so hopefully you won’t need to either. Have a look at my complete Excel example here:
https://batchloaf.wordpress.com/2013/01/07/running-commandcam-from-excel-using-vba/
Hopefully you can adapt the example to do what you need in Access.
Hi Ted,
I’m either extremly stupid or there is something wrong with my computer. Below is the code I have, I slight modification to what you gave me:
Private Sub Command0_Click()
On Error Resume Next
Dim RetVal
Dim img, CommandCam As String
img = Application.CurrentProject.Path & “\image.bmp”
CommandCam = Application.CurrentProject.Path & “\CommandCam.exe”
‘If the image file already exits, delete it
Kill (img)
‘check to be sure it really is deleted
While Dir(img) > “”
Wend
‘take a new picture
RetVal = Shell(CommandCam, vbNormalFocus)
End Sub
At this point, I just want to go past taking the image, then I can load it into my application which is the easy part. But nothing appears. The camera light flickers as usual but there is no image. When go back to the app folder and double click CommandCam it runs and takes the picture and saves it as image.bmp. So I really dont get why it does not work when shelled, which is basically the same as double clicking the executable from the app folder. Sorry to be a bother but this is the most frustrating 2 days I ever had becuase from all thr help you’ve given, it should work but it simply doesnt! Am still missing something?
Percy.
Hi Percy,
I’m sorry to hear you’re having such a frustrating time with this. I doubt that your computer’s broken and I’m sure you’re not stupid! My instinct tells me that your problem is something to do with VBA accessing the wrong directory (or directories).
Firstly, some suggestions:
If that doesn’t solve the problem, then I have a couple more questions:
Ted
Ted, you are a genius!! And thanks a million for the patience. After trying so many little tweaks and adding a number of references and removing some later, it works now! I couldnt belive it, the first time it worked cos I really dont know what I did but its working perfectly now.
Thanks again.
Great, well done Percy! Best of luck with the rest of your project.
Ted
Hello!
First I have to congratulate you for this application. It really works fine. 🙂
Second: I have a problem and it would be awesome if you can help me solve it 😛 I don’t know why, but the image that comes from my webcam is a little bit darker. I guess if I can turn on the night mode before I take the snap I could solve the problem.
So the problem now is to turn on the night mode. I was researching on msdn website and I found this topic (http://msdn.microsoft.com/en-us/library/windows/desktop/dd387908(v=vs.85).aspx) with some functions that can change the brightness, contrast and other settings, but I can’t find anyone that can turn on the night mode.
Do you now how to do that or something else that can solve my problem? That would be a very big help.
Tanks 🙂
Hi Fernando,
Thanks for your message. Have you tried adding in a short delay before CommandCam takes the photo. For example, to include a 5 second delay before taking the snapshot,
Sometimes, a delay like this can allow the camera to “warm up” and you get a clearer picture.
Unfortunately, I don’t know what “night mode” is exactly (although I can imagine what it does of course) – there’s no setting like that on my webcam. Many camera drivers provide a custom control panel that lets you adjust settings specific to that camera – these camera-specific control panels can usually be accessed through the Cameras section of the Windows Control Panel; just find where your camera is listed and right-click on it. Some of these control panels allow you to configure default settings for the camera. If that’s the case for your camera, it could help since “night mode” is probably something specific to your model of camera, unlike for example “brightness” which is probably adjustable on most cameras.
If changing the brightness setting would solve your problem, it looks like the IAMVideoProcAmp interface you found on MSDN might do the trick. To add it into the code to solve your problem mightn’t be too bad (i.e. setting the brightness to one specific value that works for your camera), but adding it to CommandCam as a general feature would involve more work than I have time for right now 😦
Anyway, before doing anything more complicated, try adding the delay as I suggested above (if you haven’t tried that already) and let me know how you get on.
Hello. Tanks for your replay 🙂
I have tried the delay and the result was the same.
I also already implemented the brightness with IAMVideoProcAmp interface in you code but that was not what I want, because the brightness seems like a virtual increase of brightness made by software, but the “night mode” that I’m talking about seems like a physic thing, like if the lens of the webcam open to let more light enter or something. I know that because the software that comes with my pc to control webcam has that functionality and if I turn the night mode on in the software and take a snapshot with CommandCam the result comes perfect. So what I wanted to know was if there are a way to do that in the code, so I don’t have to open the webcam software every time before use CommandCam.
Tank you anyway and sorry for my late replay 🙂
Hi Fernando,
What make and model of camera are you using? I’ll see if I can work out how night mode can be activated programmatically.
I’ve just done a little digging around and I did find some information that could be useful. What I think are the most likely settings to be changed for “night mode” are the camera aperture and exposure time. Both of these are controlled using the IAMCameraControl interface:
http://msdn.microsoft.com/en-us/library/windows/desktop/dd387908%28v=vs.85%29#camera_settings
It might be possible to query the IAMCameraControl interface to see what aperture (aka “iris”, I think) and exposure settings are used with and without night mode enabled (using the camera’s own configuration application to turn it on and off).
Regards,
Ted
Hi Ted,
That’s a good idea, but I already tried to use the IAMCameraControl and it don’t work with my webcam (or I’m doing something wrong, but I don’t think so, because I used the IAMVideoProcAmp just fine and the way both works should be essentially the same). What happens it that when I use the GetRange method it don’t succeed to return in any property. So, or I’m doing something wrong, or my webcam don’t support any property of IAMCameraControl interface.
This is the adaptation that I made to the MSDN IAMVideoProcAmp example code to work with IAMCameraControl:
IAMCameraControl *pProcAmp = 0;
hr = pCap->QueryInterface(IID_IAMCameraControl, (void**)&pProcAmp);
FILE *cam = stdout; // just some file I was using to store the printed messages
if (FAILED(hr))
{
fprintf(cam, “QueryInterface Error\n”);
}
else
{
fprintf(cam, “QueryInterface OK\n”);
long Min, Max, Step, Default, Flags, Val;
hr = pProcAmp->GetRange(CameraControl_Pan, &Min, &Max, &Step, &Default, &Flags);
if (SUCCEEDED(hr))
fprintf(cam, “CameraControl_Pan\n”);
hr = pProcAmp->GetRange(CameraControl_Tilt, &Min, &Max, &Step, &Default, &Flags);
if (SUCCEEDED(hr))
fprintf(cam, “CameraControl_Tilt\n”);
hr = pProcAmp->GetRange(CameraControl_Roll, &Min, &Max, &Step, &Default, &Flags);
if (SUCCEEDED(hr))
fprintf(cam, “CameraControl_Roll\n”);
hr = pProcAmp->GetRange(CameraControl_Zoom, &Min, &Max, &Step, &Default, &Flags);
if (SUCCEEDED(hr))
fprintf(cam, “CameraControl_Zoom\n”);
hr = pProcAmp->GetRange(CameraControl_Exposure, &Min, &Max, &Step, &Default, &Flags);
if (SUCCEEDED(hr))
fprintf(cam, “CameraControl_Exposure\n”);
hr = pProcAmp->GetRange(CameraControl_Iris, &Min, &Max, &Step, &Default, &Flags);
if (SUCCEEDED(hr))
fprintf(cam, “CameraControl_Iris\n”);
hr = pProcAmp->GetRange(CameraControl_Focus, &Min, &Max, &Step, &Default, &Flags);
if (SUCCEEDED(hr))
fprintf(cam, “CameraControl_Focus\n”);
}
fclose(cam);
This code should print the properties that return successfully but it only prints “QueryInterface OK”. So I think this means that none of the properties are supported.
My webcam is a Chicony CNF7051.
Tanks Ted, I’m really appreciating your help 🙂
And sorry for my English 😛
Hi Fernando,
Perhaps it would be worth printing out the hr values returned from each of those function calls just to be sure what’s stopping it from working? For example,
Once you’ve printed them out, you can look up the hr values here:
http://blogs.msdn.com/b/eldar/archive/2007/04/03/a-lot-of-hresult-codes.aspx
Ted
Hi Ted,
I done what you said. The error is:
hr = -2147023728
But I can’t find the meaning of the error anywhere.
Hello again Fernando,
Sorry, I think that was my fault – I should have said to print the error code as unsigned int or hex (“%u” or “%x” rather than “%d”). Anyway, I’ve converted the number you got into an unsigned int and then printed it as a hex value – 0x80070490. I then googled “DirectShow 0x80070490” and I found the following page:
http://msdn.microsoft.com/en-us/library/windows/desktop/dd375623%28v=vs.85%29.aspx
I did a search on that page for the error code you received (in hex format) and I found this:
So, unfortunately that doesn’t sound good for controlling your camera this way! We’ll have to try to think of a different way.
Ted
Hello Ted,
Well, seems like that the problem is with my camera.
I think I already gave you much work, so I will try to find a solution for may camera on my own and if I succeed I will let you know.
Tank you for everything and once again congratulations for the excellent software you have made 🙂
Hi Fernando,
Ok, best of luck getting it working. One final tip, by the way: In case you haven’t seen this:
http://forums.computers.toshiba-europe.com/forums/thread.jspa?threadID=39643
It sounds from the last message in that discussion that the drivers for at least some Chicony cameras may have been updated to remember the night mode setting. If that’s the case for your camera, then updating your drivers might solve the problem.
If not, best of luck finding another solution!
Ted
oraleee: muy buen trabajo.
probado en win7 y compilación exitosa en vs 2010.
Hola Alfredo,
Muy bien, gracias por tu mensaje. Estoy encantado de saber que está trabajando con VS2010 y en Windows 7!
Ted
Dr. Burke, if command cam could select a camera not by type but by serial number as seen in the USB device entry, it would be a big benefit to me. Please see email sent…
Hi Ken,
Indeed I think it might be possible to do more or less as you suggest. Each capture device has a unique string called its “DevicePath”. My understanding is that this string can be accessed in the same way as the device name is accessed on lines 274 and 326, e.g.
pPropBag->Read(L”DevicePath”, &var, 0);
I’m not on a Windows PC here, so I won’t be able to try this until I’m back in the office, but if I do find a few spare minutes over the next few days, I’ll see what I can do.
Regards,
Ted
Thank you. I think the camera supplier fed me a line about their software. As I wrote in my other communique, I have two of the same device id cameras on the same computer so the serial number is the means of accessing individuals.
Yes, I got the following info from MSDN (http://msdn.microsoft.com/en-us/library/windows/desktop/dd377566%28v=vs.85%29.aspx):
…so it sounds like that will do the trick for your two identical cameras.
I may be speaking stupidly to you, but the NIRSOFT USBDEVIEW software shows this Serial Number field…
Serial number is desirable since the device can be plugged in to the USB facility of the computer in different slots…
According to this discussion on MSDN, it sounds like the DevicePath string will either be set to the serial number, or that it will contain it as a sub-string. In either case, once the DevicePath is obtained, it should be possible to match it easily (i.e. in a single line of code) to the serial number in a way that works when the device is plugged into any port.
Excellent!
Hi Ken,
I’ve added in a new feature for selecting the camera by serial number. To specify a camera with the serial number “314159265”, just type:
This feature works by searching the DevicePath for any occurence of the specified serial number as a substring. You can now view the full DevicePath string for each attached device by typing:
Let me know if this works for your two cameras.
Regards,
Ted
Thank you Ted. Unfortunately, the serial number seen by NIRSOFT USBDEVIEW is not present in the list of devices. Only the Device Name is present. Here is the result…
I cannot get the new option to work. I wonder what overnight shipping for my camera is to Ireland?
“CommandCam /devserial 30210152”
“Video capture device not found”
Hi Ken,
Please plug in both cameras and run “CommandCam.exe /devlistdetail”, then reply here with the full output that appears in the console. I want to see if there are differences in the two DevicePath strings apart from the ones relating to the different USB sockets.
Ted
If you can get your code to see the serial number of an Arduino Uno R3, I think you will have it made. The serial number seems to appear in the same field.
Available capture devices:
Capture device 1:
Device name: FJ Camera
Device path: \\?\usb#vid_04f2&pid_b186&mi_00#7&e3c804d&1&0000#{65e8773d-8f56-1
1d0-a3b9-00a0c9223196}\global
Capture device 2:
Device name: DFM 72BUC02-ML
Device path: \\?\usb#vid_199e&pid_8207&mi_00#7&98e30b1&0&0000#{65e8773d-8f56-1
1d0-a3b9-00a0c9223196}\global
I only have one DFM… camera here now. Looking at the string, I don’t see the serial number seen by other software. The manufacturer of the camera makes stuff than can see this, but the software isn’t command line based. Useless to me!
Ah right, sorry, I only saw your second message just now. I’ll tell you what: once you have the two identical cameras in the same place, let’s try the same thing again (“CommandCam.exe /devlistdetail”). Hopefully the two identical cameras will have DevicePath strings that will allow them to be distinguished.
Ted
Hi Ken,
I’m a bit confused. These two cameras are clearly not the same model at all! They have different friendly names (“FJ Camera” and “DFM 72BUC02-ML”) and different vendor and product IDs. There should be no problem whatsoever selecting either one of these cameras consistently using its friendly name – there’s no need to resort to serial numbers!
For example, for the first camera:
or for the second camera:
Am I missing the point here somehow?
Ted
Yes sir, they are not the same. But I am looking at a configuration that I will see in the near future that will have two DFM… cameras.
I do not have the second camera yet. I always try to prevent a disaster before it happens.
Ok, well the next best thing you can try is the following:
Plug the single camera into a USB socket
Run “CommandCam.exe /devlistdetail”
Now, unplug the camera and plug it into a different USB socket.
Run “CommandCam.exe /devlistdetail” again
Copy the output from both times into a reply here so that I can compare them.
I should be able to see from the output whether this approach will work. In fact, it would be ideal to try the camera in more than two USB sockets just to be sure. If each physical camera is uniquely identifiable, then a certain part of the DevicePath will stay the same as the camera moves from socket to socket. However, if Windows cannot distinguish two cameras of the same make and model, it uses the USB hub and port to generate this part of the DevicePath which means that it changes when the same camera moves around.
Dr. Burke, I have tried that! Results! The one part that seems constant and seems as though it can be used as a serial is the “98e30b1”. Indeed, a few minutes ago, I tried this number and it will acquire an image with this number. But, the hex value of this number seems to be quite different than the number that other software sees.
Position 1:
Capture device 2:
Device name: DFM 72BUC02-ML
Device path: \\?\usb#vid_199e&pid_8207&mi_00#7&98e30b1&0&0000#{65e8773d-8f56-1
1d0-a3b9-00a0c9223196}\global
Position 2:
Capture device 2:
Device name: DFM 72BUC02-ML
Device path: \\?\usb#vid_199e&pid_8207&mi_00#7&98e30b1&0&0000#{65e8773d-8f56-1
1d0-a3b9-00a0c9223196}\global
Position 3:
Capture device 2:
Device name: DFM 72BUC02-ML
Device path: \\?\usb#vid_199e&pid_8207&mi_00#7&98e30b1&0&0000#{65e8773d-8f56-1
1d0-a3b9-00a0c9223196}\global
Position 4:
Capture device 2:
Device name: DFM 72BUC02-ML
Device path: \\?\usb#vid_199e&pid_8207&mi_00#7&98e30b1&0&0000#{65e8773d-8f56-1
1d0-a3b9-00a0c9223196}\global
To be absolutely sure, I need to get my hands on another camera ASAP.
This number according to Nirsoft is part of the “ParentId Prefix”. I need to look to make sure it will be unique. In the meantime, I am also having another camera sent here.
Okay, I’ve ordered another camera. I also tried the /devlistdetail on a different windows computer. Same “98e30b1″…
Okay, I’ve ordered another camera. I also tried the /devlistdetail on a different windows computer. Same “98e30b1″…
Ok great! Fingers crossed that the second camera will work the same (but with a different id number of course). Based on what you observed when plugging the camera into different sockets (and on different PCs), I’m very optimistic that it will work, but only time will tell for sure.
Please let me know how you get on. I’ll be delighted if it works, but if not I’ll try to think of another workaround.
Ted
Dr. Burke. I have confirmed as of a few moments ago that two cameras of the same type are individually addressable via the “ParentId Prefix” scheme you implemented. Thank you very much!
Bravo! That’s great news. Best of luck getting everything else working!
Ted
Dr. Burke, I am delighted to be working with you.
Thanks to CommandCam: http://alziende.nl/grohe/
Pingback: Gitshots für Windows : Zustandsforschung
Hello.
Excelent tool for automation, and thanks for making it opensource !
I have 1question and 2 feature requests. hehe
I have a few camera models with autofocus.
Is there any way to set the focus to be constant and disable autofocus?
The camera will be pointing always in the same direction, and autofocus adds a delay when capturing the image.
Another question.
Would it be possible, to read the parameters form an .ini file, and specify that file as the only command line parameter?
Another question.
Would it be possible, to capture múltiple devices while commandcam is launched only once ?
I think that, the OS process spawn time is what takes most of the resources while waiting to have an image. Of course we are talking of half a second per launch.
Kind regards,
Diego
Hi Diego,
Thanks for your kind comments. I’ll do my best to answer your questions.
Regarding the autofocus, I’m afraid I don’t know much about changing that setting programmatically. It may be possible to configure it using DirectShow or some other API, but I haven’t ever done it. Does the manufacturer of your camera provide a control panel that allows autofocus to be switched on and off? If not, then my guess is it may not be possible to disable it. If this setting is available though an existinng control panel for the camera, is it possible to disable it that way and then use CommandCam without the focusing delay?
You could also have a look at RobotEyez, another image grabbing program where I’ve been trying out new features for the next version of CommandCam. It allows multiple images to be taken sequentially. Since the autofocus would only happen at the beginning, it should allow images to be captured more frequently.
Regarding reading the parameters from an .ini file, what parameters do you have in mind? Is there a specific reason that this feature would be useful for you? Is it that you want to run CommandCam from a different program, but with different command line arguments each time? If so, there are ways to do this without using an .ini file. If you can explain why you want this feature, I’ll have a think about possible solutions.
Regarding capture from multiple devices: This is a feature that someone suggested previously and I would certainly like to include it, but it requires quite a bit of modification of the existing code, so it will have to wait until I have more time available to concentrate on CommandCam. I develop this software as a hobby and I have a busy teaching load here in DIT, so it’s hard to find the time during term! Hopefully, I’ll get around to it soon though – I think it would be a really useful feature for a lot of people.
Ted
Hi, thank you for the consideration !
I have already handled all of it with opencv & python.
Thanks !!
Diego
Ok, glad to hear you got your problem solved!
Hi, I’m trying to use your program, and it’s saying it’s working and saving the file image.bmp, but it’s not showing in the directory afterwards.
Hi Kevin,
It sounds like it’s probably running but saving the file in a different directory than the one you expect. Are you running CommandCam by typing in the command in a command window, or are you running it from within a batch file, or are you running it by double clicking on it?
However you’re running it, I suggest doing a file search in Windows for any files called “image.bmp” – it might just show up somewhere, which would clarify what’s happening.
Let me know how you get on.
Ted
This is pretty great. I’ve been looking for something like this for a project for a while. Thanks.
Hi Rob. Thanks for your kind comment and of course for your generous donation. It’s very much appreciated. I hope your project works out well!
This is great thanks! But is there a way to get rid of the delay time from executing the program and a file being saved. I have tried ‘CommandCam /delay 1’ but it’s still a few seconds before the image is saved. I need the capture to be instant in the application I am writing.I notice in the source code the default delay time is 2000, does this have something to do with my problem, or is it just my webcam (built in). Apart from this your application works great!
Hi Robin,
CommandCam accesses the camera as a video device and unfortunately it just seems to take some time to start it up. An additional delay can be specified since some cameras take a couple of seconds to automatically adjust the brightness level. However, even with the delay set to zero, it still seems to take a second or two to start up the camera.
I’ve been thinking about this problem and considering other solutions like providing a two program solution: one that runs in the background and keeps the camera running, and a second program that lets you just grab a snapshot (from the already open camera) at any particular moment at the drop of a hat. It seems a bit messier than the appealing simplicity of CommandCam, but maybe it’s worth considering?
Ted
This is an excellent idea Dr. Burke. I am sure it could potentially speed up operations in my application.
I am really impressed..I am using windows 7 and when I try to run it as WINDOWS task scheduler, Its activating the camera as lights do blink,but I don’t see the picture getting saved…help me
Hmmm, interesting problem. If you run it with the task scheduler, I’m not sure what directory the image file will be saved in by default. Have you tried using Windows file search to find the file?
Alternatively, you could try using CommandCam’s “/filename” switch to specify the full path of the output file. For example:
CommandCam /filename "c:\output.bmp"
Hopefully that will save the file in the root directory of the C drive, but you could specify whatever directory you want.
My finding – running this through Task Scheduler saves the file by default to: C:\Windows\SysWOW64
Thanks Justin, that’s a useful thing to know.
Ted
I was pretty excited to have found this at first, but seem to have found a glitch. When I try the code this way, first: commandcam /devnum 1 /filename c:\test.bmp
it works. The second time i re-run the same line and it it returns an error of “Could not render capture video stream”. So then I tried commandcam /devnum 2 /filename c:\test.bmp, and it works using the front camera; and continues to work multiple times using the front camera. The rear camera; the one I really want to use, will only work once, until system reboot. I have to restart the computer for it to ever re-capture from the devnum 1; and only have it capture once, followed by the error message I’ve specified. Very annoying. Why is this?
Hi Robert,
I haven’t come across this problem before, but I don’t have a 64-bit Windows computer to try it on. Perhaps someone else can comment if they’ve run into a similar problem.
As a matter of interest, did you try opening the cameras in the opposite order? If you open the front one first and then the back one, maybe you’ll be able to keep opening the back one?
Finally, I have another program called RobotEyez, which will probably become the next version of CommandCam when I get around to finishing it. In the meantime, you can try it – it might just do what you want. Here’s the link:
https://batchloaf.wordpress.com/2011/11/27/ultra-simple-machine-vision-in-c-with-roboteyes/
Please let me know how you get on.
Ted
oh and; windows surface pro 64bit….2 cameras. I want the rear Camera to work many times over with yur app. only woks with the front though.
Dr. Burke. Your commandcam program is wonderful. I have told you that before. I asked you to add the feature for addressing singly among multiple cameras of the same type. You did that. I commend you. I have conquered every physical problem with lighting, camera settings, LCD segment analysis software, creep of plastic parts around fine adjust shims holding the camera…. Now, I have a problem with the manufacturers drivers on my customer’s WIN XP on the CORE 2 DUO system with 2G or storage and directx9c causing hangs one or more cameras, when I have 4 plugged in. WIN7, i5, 6G… Not a problem. It behaves strange with the suppliers GUI app as well when there are 4 under XP. I see strange hanging, slow responses with the supplier software as well as commandcam. Seems like a driver problem too me?
Hi Ken,
If the computer is hanging, then it probably is a driver issue, since a normal application shouldn’t have enough OS privileges to do that. Also, if you’re seeing the problem when using the supplier’s GUI, then it lends further weight to that theory.
Is it an option to try different cameras instead? I could be wrong, but I’d imagine 4 simultaneous USB cameras in Windows XP is not something many people have done, so the manufacturer probably wouldn’t have done much testing under those circustances. What kind of cameras are they?
Ted
The Imaging Source. DFM 72BUC02-ML They are not being used intensively. One camera is queried 2 times per minute. Then, the next. and so on. We use the hashed Windows identifier that you added the feature into commandcam to address them. Works wonderfully most of the time. It isn’t the computer that hangs, it is the task and camera that hang. Control-Break and unplug/replug. Then it comes back to life. One in a group of 4 is always at a 50/50 chance for non operation. We went to a better hub and when they operate, they operate much faster. But still some hang. We went to “native” attach to the core 2 duo, XP, system and the behavior is actually slower with commandcam and CC capture. I called Germany from MX and spoke to a support person there. Of course, it is the other guy’s software. But this even happens at low frame rates with their GUI tool. They are sending me a different command line tool. I have to get this running by Saturday or they’ll send the Zeta cartel after me…
Hi Ken,
I did a bit of googling, but didn’t turn up anything very useful. I’m not sure I can offer much help with this since I won’t be able to reproduce the problem here. Did you make any progress yourself?
Ted
hello sir. iam trying to use cmdcam with access form. i have a button on a form to call cmdcam.exe. everything works well except that there is no image file as output but when i double click the exe directly it works fine
I am using Shell cmd in vba to achieve this
My code in access 2010 :
Dim x as variant
x= CurrentProject.Path & “\CommandCam.exe”
Call Shell (x,vbHide)
Hi Sanju,
Unless you specify a different folder for the file to be saved in, CommandCam will place the file in the current working directory. When you run it from Access using the shell command, I’m not sure what the working directory will be.
Perhaps you could try specifying the full path in the filename that CommandCam should use. For example,
I’m not sure whether you will need to “escape” each of the backslahes if you’re putting that into a string for a shell command. Depending whether or not you need to do that, your line would probably end up looking like one of the following:
or
Alernatively, you can try changing directory before calling the shell command, as I did in this Excel VBA example:
https://batchloaf.wordpress.com/2013/01/07/running-commandcam-from-excel-using-vba/
Hopefully that helps. Let me know how you get on.
Ted
Hello, I happen to stumble upon your project and found it to be very helpful. I have so far managed to make the camera to capture a multiple images at a very fast speed, but then I came across a problem where I have more than one camera and couldn’t capture from all 3 of them. I’m trying to remake a program to capture from all 3 cameras is there a way?
Hi Parker,
A few other people have expressed interest in a multi-camera version of CommandCam, so I’m hoping to incorporate this into the next version. However, I’m not sure exactly when I’ll get around to it. I’m currently unable to do any Windows development because my Windows laptop has bitten the dust so I’m completely Linux based at the moment.
Hopefully I’ll get around to it soon though! I’m pretty sure it should be possible. Can you please give me some details about what kind of capture you need to do – i.e. how many cameras, for how many frames, and at what frame rate?
Ted
Hi, Ted
I will be using 3 cameras and ring loops will be used so infinite frames but it will be storing 2000 images per camera till it loops back and re-saves from the start, and frame rate will be around 140 fps (its a high speed cameras)
Ok, wow, that’s a lot of video! I’ll try to keep this in mind when I get back to CommandCam development. It won’t be for a little while though because my Windows machine is out of action, so I’m Linux only at the moment.
I’d really like to get CommandCam working with multiple cameras at once.
Ted
Hi,
First of all, EXCELLENT WORK! this program is just awesome. Is there any way you could change the initial delay to 100ms because when i try to “/delay 500” (or anything below 2000) it doesn’t take that into account and does 2000ms delay. Also is there any way you could disable the flash (or whatever you call that tiny white light next to webcam) from integrated webcam so that the user doesn’t know that webcam has turned on and taken a picture. Again thanks for this wonderful application
Hi Niravshah,
Thanks for your generous words. Sorry for the delay replying – I’ve been on holidays.
Unfortunately, the 2 second delay you’re experiencing is not introduced by CommandCam – it just takes Windows a couple of seconds to open the camera and wait until it’s ready to deliver data to the program that opened it. Some cameras are faster than others, so you could try a different one to see if it produces a shorter delay.
If you need a really fast response to snapshot some event, you probably need to use a different program that keeps the camera open and just records a frame instantaneously when something happens to trigger it.
Regarding the light on the camera, it’s not normally possible for software to disable it, because it’s there to let the user know that some piece of software is accessing it, for example so that malware programs cannot spy on computer users without them realising it. I suppose it would be possible to disable the light by physically disconnecting the LED or whatever, but I wouldn’t recommend doing that! Also, there may be some cameras out there that don’t have a light, but all of my USB cameras have one.
Ted
Pingback: Computer Security: Snap a Picture of Whoever Logs on Your Laptop
Hi
I tested against a “Microsoft« LifeCam VX-2000”.
that got converted to; …commandcam.exe /devname “Microsoft½ LifeCam VX-2000”
commandcam.exe replied; Video capture device Microsoft╜ LifeCam VX-2000 not found.
I can see that the device name choice may not have been the best.
Hi Terry,
That’s an interesting problem. I haven’t come across camera names with those kind of exotic characters before. Could you try running “CommandCam /devlist” or “CommandCam /devlistdetail” to see what the camera shows up as there? You should still be able to select the device by device number or by serial number, although the device number may change depending on the order cameras are plugged in or which USB socket is used for each one.
Ted
Capture device 1:
Device name: Microsoft« LifeCam VX-2000
Device path: \\?\usb#vid_045e&pid_0761&mi_00#6&2dda4885&0&0000#{65e8773d-8f56-11d0-a3b9-00a0c9223196}\global
The character is a ®. (Registered trademark)
I am using /devnum to access the webcam at the moment.
Btw. It’s a great program. Thank you!
Hello, I have proposed you for Nobel prize, and I haven’t checked your code on Acer but I would like to make a small proposition, I’m very good in VBA but C++ and other C-s are total mistery for me. If you could help me to adjust CommandCam.exe code for taking pictures with preview and naming files with date of creation (and not deleting them) I would press that DONATE button so strong it would fall of screen 🙂 . Please say yes 🙂 . p.s. Sorry my english is not very good but I’m trying 🙂 .
Hi Tom,
As it happens, I may have an hour or two free this afternoon to do a little programming. Can you explain a little bit more about how you want this to work? So far, what I think you’re looking for is:
1. Run CommandCam from the command line. (Or do you want to run it from Excel VBA?)
2. Display a preview window showing the moving image from the camera.
3. Press a button or something to capture an image.
4. Save the image to a file using the date and time as the filename.
Is that correct? If so, please give me an example filename, so that I can format it the way you need it. I’ll have a look at this later, but no guarantees! By the way, if I do get it working, there’s no need to donate. Of course it’s nice when people do, but it’s almost as nice when people just say thanks. I’m not really in it for the money, which is just as well – donations are few and far between 😉
Ted
If you get it to work it would be my honor to donate some money. File name is easy INS_REP_datetime.jpg. I would like to run it from Excel VBA and to have preview in workbook. And to be able to Press a button or something to capture an image. Thanks for your time. I appreciate it.
If you get it to work it would be my honor to donate some money. File name is easy INS_REP_datetime.jpg. I would like to run it from Excel VBA and to have preview in workbook. And to be able to Press a button or something to capture an image. And one more thing. Not to delete pictures that were created before. Thanks for your time. I appreciate it.
I have managed to solve all problems I had with integration of commandcam.exe into VBA. If you want I can send you sample workbook. Thanks 🙂
Thanks …this post helped me a lot…
Thanks Sarthak!
Hey, I want to trigger camera & save image in a particular location I have only VBS to trigger your programm. Can anyboday share a script for it to me.
Thanks
Hi Arun,
Have you looked at this example?
https://batchloaf.wordpress.com/2013/01/07/running-commandcam-from-excel-using-vba/
Perhaps it will help.
Ted
Pingback: Windows – Mac – Android – IOS – Graphic – Wallpaper – Webmaster | Topdl Computer Security: Snap a Picture of Whoever Logs on Your Laptop - Windows - Mac - Android - IOS - Graphic - Wallpaper - Webmaster | Topdl
Pingback: Windows – Take Webcam Picture At Set Intervals | Project Failure
Where are the pictures saved to? Can we change the location?
By default, the pictures are saved to the current working directory. However you should be able to specify whatever location you want using CommandCam’s “filename” command line option. For example,
That should save the file in the root directory of the C drive, but you could specify whatever directory you want.
Great job! Solved the issue on file name by using some lines in a batch file:
set d=%date:~-4,4%%date:~-7,2%%date:~0,2%
set d=%d: =_%
set t=%time:~0,2%%time:~3,2%%time:~6,2%
set t=%t: =0%
RENAME “image.bmp” “image_%d%_%t%.bmp”
Works awesome! The occassional “all-green” photo happens but I assume that’s a flaw of the camera.
Thanks Ted!
Thanks Pichu, glad you found it useful!
Thanks too for that great tip about naming the files with the date and time.
Ted
Hi Ted.
All I have to say is “What a wonderfull work” and thanks for answering all questions and comments, not everyone does it.
I am going to trate to use it on Web Applications. I’ll talk you about results.
Congratulations!!
Thanks!
Ted
Pingback: Gitshots für Windows : Zustandsforschung
Hi,
Thanks for a great code.
Did you introduce the resolution change to CommandCam yet ?
Also when i execute the preview command the preview screen only comes up for 2 seconds then closes, can i keep it open for longer ?
Hi, i`ve made it via bat file:
CommandCam.exe /preview /delay 10000
Hello Ted! Great application. The best choice for our Citrix XenApp environment. The same question: is there an option to change resolution in commandcam?
Hi Extremely Toxic,
I haven’t changed anything in CommandCam for a while due to a lack of free time. However, you might find that RobotEyez (another program I wrote previously) does what you need:
https://batchloaf.wordpress.com/2011/11/27/ultra-simple-machine-vision-in-c-with-roboteyes/
It allows you to specify the resolution.
Ted
Ted, roboteyez is great too, but it can’t render the video stream in xenapp session. For me it’s critical. Is it real to make comcam with fixed resolution? For example 640*480. Tried to setup native logitech driver on the thinclient -no result. The main idea is to use the native win7 driver, cause it is redirected correctly with most of webcams. In any case, wait for donate after 15 july 🙂
Hi Extremely Toxic,
There’s no need at all to donate, but thanks for offering.
Do you know the specific resolution that you want to capture at? It’s a while since I’ve done any work on CommandCam, so looking at it now I’m struggling to remember the exact process for configuring and initiating video capture. However, as far as I can tell it just captures at whatever the default resolution for the camera is, which may be different for each camera. Furthermore, it may be possible to modify the default resolution using a camera-specific application (a camera properties dialog box or something like that). You could try checking in the Windows Device Manager to see if such a configuration utility is available for your camera.
Also, have you got any idea why RobotEyez won’t work in a xenapp session although CommandCam will?
Ted
Hi again. I have no idea about RobotEyez rendering. Perhaps this problem connected with the OS version. (Works on Win7x64/does not work on Win2k8r2). Tried to change the Citrix HDX policies – no result.
Yep, i`ve tried to change the default resolution on local PC. No result.
Probably i have 10000+ users who need to use over 10 different webcam models in XenApp session. CommandCam works with 90% of webcams, and is an universal solution for all users. The only trouble is height and width change availability.
P.S. Replied here cause there is no “Reply” button after your last post.
Hi again,
Ok, I’ll have a look at this later this afternoon to see if I can spot a way to set the resolution in CommandCam. I know that when I looked at this before, it turned out to be surprisingly awkward, which was one of the reasons I began experimenting with a completely now program (RobotEyez). However, since RobotEyez isn’t working for you, I’ll have another go at setting the resolution in CommandCam. It’ll have to wait until later on though because I’m working at home on my Linux laptop as we speak.
I may need to get you to test it though because I won’t be sure just by running it on my own machine.
Ok, I’ll get back to you later.
Ted
P.P.S. Specific resolution is 640×480.
Great thanks! Waiting %)
Hi Extremely Toxic,
Just a quick update. No success so far, I’m afraid. I spent a bit of time yesterday trying to work out how to set the resolution for the DirectShow sample grabber filter, but it’s surprisingly complicated. It’s becoming clearer to me why I started over with a slightly different approach with RobotEyez!
I haven’t given up on getting this to work, but I can’t give any guarantees and I’ll be going away for the weekend tomorrow, so if I don’t get it working today it could be next week before I get to look at it again.
Ted
Hi. Thanks for this application. It`s great for our Citrix XenApp environment. The same question about resolution change.
I use http://code.google.com/p/jpeg-compressor/ to convert the files to jpeg.
Another happy user! I have a weather station at an airfield and run a simple website using cumulus. I need to take 4 pix every 30 minutes so folk can see what the sky looks like ….. I broke my head trying to connect 4 cams to one USB hub to one computer. Everything I tried gave up after one camera.
This just eats the job! I even did a donation 🙂
Looks like http://www.easy2convert.com/bmp2jpg/ will handle the rest of the processing.
Ian Boag in New Zealand ……
Wow, thanks for letting me know that – what a fascinating application! I’m delighted to hear that CommandCam did what you needed. Thanks also for the generous donation – it’s much appreciated!
Ted
This program is awesome and exactly what I was looking for. I am wondering how hard it would be to change the output to PPM (pixel map) format. I also wish the resolution was selectable. For me it might be desirable to have a smaller resolution to keep the file sizes smaller (320×240). Again Thanks for this very useful program!
Hi Mark,
Would you consider installing another utility to convert the image after it’s captured? For example, I normally use ImageMagick, which is free to download and install, and contains numerous useful utilities for command line image processing and conversion. Imagemagick is incredibly useful and can be used to automate all kinds of image-related jobs.
You can download the Windows version of ImageMagick here.
Assuming you’re running CommandCam in a batch file, you would simply add an ImageMagick convert command straight after the CommandCam line, something like the following…
ImageMagick is actually a suite of (mostly command line) software utilities, of which “convert” is just one program. So it’s the second line in the example above that actually runs ImageMagick.
You can also use ImageMagick to reduce the resolution of the image, so that if you’re archiving a lot of images they will use up less space. Something like…
The “-resample” operation is just one of a long list of image processing operations that ImageMagick can perform. For example, see the long list of options for the convert command.
If you want each archived image to be saved with a filename that includes the current time, you could try something like…
If you’re trying to save storage space, you might also want to consider using a more compressed image format, such as jpeg…
WARNING: I wasn’t able to test any of these examples because I’m working on a linux laptop here, so apologies if some of them require modification to make them work correctly!
Anyway, would something like the above ImageMagick examples do the trick in your application?
Another option you could consider is to try RobotEyez, which is another program I wrote a couple of years ago for automated image capture. It provides some additional options not supported by CommandCam.
Ted
Ted, thanks for all of this and I will give it a try. Please ignore the bonehead post I made below. This is a project that I do not have to pleasure of working on full time. I come back to it on occasion and I had forgotten I had asked you about pixel map format. Hope I can get it to work. Thanks again!
Mark
Hi Mark,
No problem at all. My recommendation is still basically the same though – use ImageMagick. It can handle conversion between just about any two image formats and does lots more besides. It’s free software and only takes a minute to install.
Hope you get things working!
Ted
Can’t get the object to be found, keep getting a 424 error. Any ideas?
Hi Chris,
Can you be a bit more specific? Are you seeing this error when you run CommandCam or when you try to compile it? If it’s the former, please explain a bit about your setup and include the exact command line you’re running. If it’s the latter, please explain which compiler you’re using and tell me if you’ve modified the original code in any way.
Either way, it would also be useful to know what version of Windows you’re running.
Ted
Dear Ted,
If i want to using CommandCam code to modify PIN_CATEGORY_STILL(real high resolution) form my camera.
hr = pBuilder->RenderStream(
&PIN_CATEGORY_STILL, &MEDIATYPE_Video,
pCap, pSampleGrabberFilter, pNullRenderer);
I want to high resolution still capture from my camera, not save a frame.
How can i modify it.
I reference MSDN , it say can use PIN_CATEGORY_STILL.
But i add it in code, the hr error code return is fail.
I have no idea how to do @@
please give me a advising.
Thanks.
Hi Alexis,
I’m afraid I can’t offer much help with this. I tried doing it myself previously, but did not succeed. As far as I could tell, most people capturing still images this way used a different API called WIA (Windows Image Acquisition) which is totally different from DirectShow.
Anyway, best of luck trying to get it working!
Ted
First time I have gotten something like this to work. Excellent program! I am running this through the Windows Task Scheduler, and I was wondering if you could tell me how to use date/time variables in the filename. Thanks!
Hi Om3ga73,
Glad to hear you found it useful!
Please have a look at Pichu Frisby’s comment above:
https://batchloaf.wordpress.com/commandcam/#comment-3082
Maybe that will point you in the right direction? If you want to run it from the task scheduler, presumably you’ll want to include commands similar to what Pichu used in a batchfile that you can run to grad a photo with the current date/time in the filename.
Ted
Is it possible modify the program with the option for change resolution of webcam ?
Hi Massimo,
Actually, due to the way CommandCam accesses the camera via Microsoft’s DirectShow API, it’s quite difficult to set the resolution to anything other than the default. With some cameras, it may be possible to use another application (a device-specific configuration panel, for example) to change the default resolution of the camera and then use CommandCam to capture snapshots, but unfortunately CommandCam does not provide a command-line option to specify the resolution.
You could also consider trying RobotEyez which is another image capture program that I published here some time ago. As it happens, one of the main reasons I began developing RobotEyez was to get around this restriction with CommandCam. Here’s the link to RobotEyez – it’s just possible that it might do what you want:
https://batchloaf.wordpress.com/2011/11/27/ultra-simple-machine-vision-in-c-with-roboteyes/
Hope that helps!
Ted
Ted
thanks for quick answer.
I will give a look to roboteyes.
Best regard
Massimo
Ted, are you still responding to post here?
Ted, I see you are responding to posts here. I see that changing the resolution is not possible but what about changing the output format? I need pixel map format. Any suggestions?
Ted, I see I already asked you this and you answered me already ; ( Sorry to be a pest.
Ted,
Maybe I have a legitimate question here ; ). My project is detecting LEDs. I was planning to do some sort of image processing on RGB codes. That’s why I wanted PPM format because the RGB codes are pretty easy to access. Do you have any experience in this area and are there any apps out there that can process images (detect differences). Thanks!
Mark
Ted,
Terrific program and thank you for making it available. I quickly read through the thread to see if my question had been asked already. So forgive me if it has already been asked and I just overlooked it. With that said I was wondering if it is possible for CommandCam to perform a live video feed while generating an image file on command? Basically, using excel vba I would like to display a video feed from my webcam and then upon command or a push of a control button obtain a still image of the video feed without interrupting the video feed. I have already tried the x360 Video Capture Activex control to do this which can be found at http://www.x360soft.com, worked beautifully in the “trail version”, but upon paying for and downloading the licensed version I have ran into many problems and now think it might be a scam, foolish mistake on my part. So back online to look for more answers/solutions I went and found CommandCam. I believe I can make CommandCam work for me if the above mentioned functionality is not possible, but it would be extremely sweet if that functionality was possible.
Thanks in advance for your time,
Jeremy
i may seem quite stupid but how do i make it not come up with the copyright stuff? i know it says CommandCam /quiet but where do i put that? thanks, programming noob
You would type “CommandCam /quiet” in a command window. I’m not sure if you know what I mean by “command window”, so in case not, what I mean is click the start menu, select “Run…”, type “cmd” and press return. It should open a black text window where you can type in commands. That’s where you type in “CommandCam /quiet”, although you probably first need to navigate to the folder that contains the file “CommandCam.exe”. If you’re not sure how to do that, you’ll need to google “command line windows” or something like that. Specifically, you’ll need to find out about the “cd” command to change directory.
If you want to run CommandCam with the “/quiet” option from a graphical icon, rather than by typing in a command, probably the easiest way to do it is to by creating a shortcut on your desktop and setting the command to run CommandCam. For example, if your CommandCam.exe was in a folder called “C:\myprogs\”, you could create a desktop shortcut with the target set to “C:\myprogs\CommandCam.exe /quiet”.
Hope that helps.
Ted
Thank you, thanks for this software! Can’t believe you’re not charging for it. Thinking about donating in the near future 🙂
sent donation, you’re great !
massimo from Italy
PS: if I pay, are you able to compile it as ActiveX/COM+ DLL, maybe releasing it as public domain or LGPL or MPL, so it can be called from VBScript?
Hi Massimo,
Thanks! Sorry, but I’m not sure how to compile it as an ActiveX/COM+ DLL. However, the code is available under GPL, so perhaps someone else can adapt it for you? Alternatively, you can probably just run it as a child process from your VB script using the Shell command. Have you looked at the following page?
https://batchloaf.wordpress.com/2013/01/07/running-commandcam-from-excel-using-vba/
Perhaps that is of some use to you?
Ted
Hi,
First of all thanks for CommandCam.exe
I am not a real C programmer, but I can read,compile and modify.
I found a couple of minor issues:
For me/my environment, the version of 24 januari 2013 does not work.
The problem is that “device_serial” is modified by some library routine, and your program uses that modified value to compare, so I always get the first camera in the list.
A workaround is displayed in the next few lines, a real fix is probably to have a separate namespace?
global search and replace:
change “device_serial” into “device_sernum”
line 109:
Add “strcpy(device_sernum, “”);
For you to fix:
The version available on github is older than the version on wordpress. It took me more than a couple of minutes to figure out why /devserial didn’t work for me 🙂
As a suggestion:
CommandCam /quiet /devlist
doesn’t produce output. I am asking for it, so IMHO “if (!quiet)” can go on lines 329, 332, 339.
Forget my previous entry, at least the part where I suggest to change device_serial.
I found the problem: it’s a buffer overflow:
CommandCam Copyright (C) 2012-2013 Ted Burke
This program comes with ABSOLUTELY NO WARRANTY;
This is free software, and you are welcome to
redistribute it under certain conditions;
See the GNU General Public License v3,
https://batchloaf.wordpress.com/CommandCam
This version 24-1-2013
DEBUG: device_serial=ThisStringDoesNotExist
DEBUG: char_buffer=713x BDA Analog Capture
DEBUG: &device_serial=0x002EF694
DEBUG: &char_buffer=0x002EF62C
DEBUG: now do: sprintf(char_buffer, “%ls”, var.bstrVal);
DEBUG: device_serial=}\{bbefb6c7-2fc4-4139-bb8b-a58bba72400b}
DEBUG: char_buffer=\\?\pci#ven_1131&dev_7133&subsys_001016be&rev_d1#4&15c8e83a&0&5899#{65e8773d-8f56-11d0-a3b9-00a0c9223196}\{bbefb6c7-2fc4-4139-bb8b-a58bba72400b}
DEBUG: &device_serial=0x002EF694
DEBUG: &char_buffer=0x002EF62C
DEBUG: strlen(char_buffer)=144
Capture device: 713x BDA Analog Capture
Capture resolution: 320×240
Captured image to image.bmp
Notice: strlen(char_buffer)==144, but
// Other variables
char char_buffer[100];
This is correct there’s a buffer overflow causing serial numbers not to be recognizedsprintf(char_buffer, “%ls”, var.bstrVal);
just over runs the world as the devicepath is reaaally long for some devices. Anyways great software mang!
Ted, this is an awesome piece of software. Simple but extremely effective. It does everything I’ve asked from it, especially the control between front and rear cameras on a windows 8.1 tablet. You’re a legend mate.
Thanks Ron. You’re very kind.
Ted
Error on Windows 8.1 Atom PC
I got the error LIBTBD; Data is not tagged properly 6 times and i get a image with resolution 448×252. Any Solution?
Hi Baltazar,
I received a comment about a similar error with another machine vision programme I published here:
https://batchloaf.wordpress.com/2011/11/27/ultra-simple-machine-vision-in-c-with-roboteyes/#comment-4584
Unfortunately, I haven’t worked out what’s causing the error. I haven’t been able to reproduce the error myself, so it’s kind of hard to work out what’s causing it.
Ted
Works on 2 machines with Intel I5. I think that is one of two things.
The processor: Intel Atom Or the camera driver works different… (Although in Skype works). Thanks
Hi Baltazar,
Yes, I did some googling myself for that error and all the results I found (there weren’t many actually) seemed to be related to drivers, so I’m inclined to suspect it’s somehow related to the specific camera driver. I couldn’t immediately see what it is that CommandCam’s doing to raise the error though. If you find out any more, please let me know.
Ted
your RobotEyez program takes much larger, clearer images, but it seems they are in B&W. Is this by design? Any way to snap in color?
Hi tokenwizard,
Because I wrote it for really simple machine vision applications my students were working on, I made the default image format PGM which is greyscale only and stores all pixels value as plain text ascii which is easy to read using stdio functions like scanf.
Anyway, you can select BMP image format by specifying the “/bmp” command line switch. For example
…which (if I remember rightly) should take a 320×240 colour bitmap after a 1 second delay.
Hope that helps.
Ted
Hi Ted,
is there a way to use CommandCam, when my cam is being used by another application? In this case i have this error: “Could not render capture video stream”.
Hi Andrea,
No, that’s probably not possible. As far as I know, Windows will only let one program at a a time open a video capture device. It’s possible that there is some kind of software video splitter out there somewhere which can open a camera and then present itself to other programs as multiple video capture devices, but I’ve never used one. You can definitely get splitters like that for audio capture, so I’m just guessing that someone might have done the same for video. DirectShow certainly allows you to split a video stream into two different streams, but only within one program as far as I know.
Depending on what software you are already accessing the camera with, maybe you could just use a screen capture tool to take a snapshot of the video? i.e. If the live video is displayed on the screen, maybe a screen capture approach will be able to take a picture of what’s on the screen. Would that work?
Ted
There are many cam sharing apps… WebCamSplitter by Verysoft is the one that I use most frequently. ManyCam is another. All of the cam sharing apps create a virtual camera that acts as a proxy for the physical camera. Instead of selecting the physical camera in your app, you’d select the virtual camera.
Hi Ted,
Why I got an error message – Unrecognised option: /quiet ?
Thanks for your code by the way!
Hi Qing,
I’m not sure why your seeing that error. Can you please post the complete command line that you’re using to run CommandCam so that I can try to replicate the error.
Ted
hello I apologize for my bad english
how can I Change the front camera with behind camera
I need this for a VBA Excel my code looks like this
Private Sub CommandButton1_Click()
Dim RetVal, base, i, picName, picPath, picNumber
Tabelle2.Cells(1, 1) = Tabelle2.Cells(1, 1) + 1
picNumber = Tabelle2.Cells(1, 1)
fotoNumber = Me.txtNumriSerik.Value
picName = fotoNumber & “-” & picNumber & “.bmp”
ChDir (ActiveWorkbook.Path)
base = ActiveWorkbook.Path & “\”
picPath = base & picName
RetVal = Shell(base & “CommandCam.exe /filename “”” & picPath & “”””, vbHide)
For i = 0 To 5
Aviso “Eapere…” & i
Application.Wait (Now + TimeValue(“00:00:01”))
Next
Aviso “Listo!!!”
ArchivoIMG = picPath
UserForm1.fotografia.Picture = LoadPicture(ArchivoIMG)
End Sub
Hi Ton,
Sorry for the delay responding. I’ve been away on holidays and now I’m up to my eyes with other work, so I don’t have time to investigate your query. Sorry I can’t be more helpful!
For what it’s worth, I don’t think CommandCam will be able to switch between front and rear cameras unless they appear as completely separate video capture devices in windows. Maybe you could check if it’s possible to open both simultaneously using some other software. If so, then you can probably just open the rear camera in CommandCam by specifying a different device number.
Ted
Hello, I hope you can respond to this, because I would like to use this with my Access Project, but can’t get it to save the file to the Application path. I see or think it is running but the file does not save to the location? It runs fine when I run the CommandCam.exe locally (by clicking it) so it runs and saves the file as image.bmp but not when I run it through Access?
here is the code i have if anyone can help me:
It is run from a button on the form:
Dim PictureCommand As String
‘build the image capture command using roboeyez
PictureCommand = Application.CurrentProject.Path & “/”
PictureCommand = PictureCommand & “CommandCam.exe /filename output.bmp /preview ”
‘MsgBox PictureCommand
Call Shell(PictureCommand, vbHide)
‘now convert and rename as needed
Dim PhotoPath As String
Dim TargetPhotoPath As String
Dim ConvertCommand As String
PhotoPath = Application.CurrentProject.Path & “\image.bmp” ‘uses the custom mydocuments function to work out where this frame image will have been saved
TargetPhotoPath = Application.CurrentProject.Path & “\image.jpg”
ConvertCommand = Application.CurrentProject.Path & “\”
ConvertCommand = ConvertCommand & “Irfanviewportable.exe ” & PhotoPath & ” /aspectratio /resample /convert=” & TargetPhotoPath
StartTime = Timer
Do Until Timer > StartTime + 4
Loop
‘MsgBox ConvertCommand
Call Shell(ConvertCommand, vbNormalFocus)
StartTime = Timer
Do Until Timer > StartTime + 3
Loop
Me.Picture = Application.CurrentProject.Path & “\image.jpg”
Form.Repaint
Hi Eggrog01,
Sorry for the delay responding. I’ve been away on holidays and now I’m up to my eyes with other work, so I don’t have time to investigate your query. Sorry I can’t be more helpful!
Ted
Hi Egrogg01,
Firstly, I think this line looks wrong…
…because you specify “image.jpg” rather than “image.bmp”.
If you’re not seeing “image.bmp” appear where you expect it to appear, it also seems likely that CommandCam is saving the file in a different location. The default location is the current working directory, but that’s not necessarily the same as “Application.CurrentProject.Path” which is presumably the project folder.
I think you should be able to specify a filename with a full path using the /filename command line option. For example,
Finally, if you want to try to work out where CommandCam is currently putting the images, ask Windows to do a search for the filename “image.bmp” and see where it finds it.
Ted
Hi Ted,
Thanks for sharing the useful tool, I really appreciate it. I tried out the software on my Win7 PC and it worked great. I then tried it on my WinXP box and can’t make it work.
It seems that no matter what I try, there are “No video devices found”. I can use the camera via it’s software, so I know its capable of working, but I’d like to take shots via shell commands. I’ve recompiled the program from your source code in hopes that error messages would give me some hints as to the problem, but no luck.
Any ideas?
Thanks,
Steve
Hi Steve,
Sorry for the delay responding. I was away on holidays and now that I’m back we’re in the thick of preparing for the Dublin Maker fair this weekend.
My first guess would be that your camera may not have DirectShow drivers for WinXP. It’s possible that the WinXP drivers which you have shown are working only support the older “Video for Windows” API, which DirectShow replaced. I’m just speculating though. If your camera is a few years old, that hypothesis would be more plausible.
Ted
Hi Ted,
Thanks for your help! I found that I had to install the “developer” version of the software that came with the camera. That seems to have installed the DirectShow drivers and CommandCam is working well now. Again, I really appreciate the handy software tool.
Steve
Hi Steve,
That’s great – I’m delighted to hear you got it sorted out!!
Ted
Ted lies. I know this because he says nobody ever clicks donate and I just sent couple pounds. Being a command line fool this util thrills me. Thanks to Mike too for turning me on to it.
Hi Rich.
You’ll be glad to hear that your lavish gift has enabled me to give up my day job and retire in luxury.
Alas… that’s just another of my trademark porky pies. Seriously though, thanks very much and I’m glad you found CommandCam useful!
Ted
PS In a desperate bid to regain the trust of CommandCam’s loyal users I’ve updated the caption on that Donate button!
Hi Ted. Thank you for posting this tool. I have a quick question. I am using this on a machine that is using a Virtual Cam Splitter. I noticed that when the app takes a picture the video feed is momentarily disrupted to the other apps consuming the virtual cam. The end result is that the other apps see a very short flash of “no video available” when commandcam is executed. Do you know if there is something in the way that commandcam is accessing the “camera” that would be causing this disruption?
Thank you!
I figured it out… it was an error with the Cam Splitter. It was blipping when another application connected.
Thanks for your little program. I was looking for an example program on how to use capture devices on Windows. I was able to compile it with mingw-w64 even without the Windows SDK. If you want I can email you the diff for the changes I needed to make.
That’s great that you were able to compile in MinGW without the Windows SDK. It’s a while since I’ve tried that, but as I recall that wasn’t previously possible. To be honest, if you sent the diff now, I wouldn’t have time to look at it because I’m flat out with teaching at the start of the new term and I’d probably just lose the file before getting a chance to try it. However, others may be interested to try compiling in MinGW, so if you want to share the diff file on google drive / dropbox and post a public link here, that would be cool.
Thanks,
Ted
Ted. I’d like to ask you about a project that I hope would be very small, if you have time. I’m in need of a windows command line tool that does nothing more than list the Video and Audio Capture Devices on the system. I’d be more than willing to compensate you for your time. Please let me know if this is something you would be interested in creating.
Hi Josh,
I figured there would already be some application out there that could do this. However, I went looking and before I found anything suitable I came across an example in C++ on StackOverflow by someone called Naveen that shows how to do it, so I just compiled that instead and it seems to work.
Here’s a link to download the executable file I got when I compiled:
https://drive.google.com/file/d/0B3NaVR72FYQccUp6NEIxc2pTc3c/view?usp=sharing
The complete code is below. Again, I must stress: this is not my code – it’s by someone called Naveen. I only changed a line or two and even those changes were almost certainly unnecessary. The original came from this page on StackOverflow:
http://stackoverflow.com/questions/4286223/how-to-get-a-list-of-video-capture-devices-web-cameras-on-windows-c
Ok, here’s the code:
I compiled it on the Visual Studio 2015 command line using the following command:
Hope that helps!
Ted
It does. The funny-ish thing was that I had found that same sample when I was researching this and, while I found it useful, I didn’t put it together that was the actual answer. It’s been a long time since I coded in c/c++ and I didn’t have an environment set up to compile it, so I may have overlooked it. Thank you for taking the time to package up a very nice answer.
You’re welcome!
Ted
Worked perfectly. The feature I could really use is a rotate command: “/rotate 90”, “/rotate 180”, “/rotate 270” since I use it on a tablet that I sometimes turn sideways. Other features others may like could be “flip” and “mirror” (if they are taking a selfie in the mirror or something).
Hi Kirk,
Would you consider using ImageMagick to do the rotation?
ImageMagick is an incredibly powerful set of command line image manipulation tools, which can almost certainly do what you want (and lots more). It’s free open source software.
Once you’ve installed ImageMagick, it would just require one additional line in your batch file to run the convert command to rotate your captured image (and optionally convert to a different file format. Something like this would probably do it:
You can read more about the ImageMagick’s convert command and the rotate operation specifically here:
http://www.imagemagick.org/script/command-line-options.php#rotate
Ted
Hello batchloaf I want to ask whether the process of capturing a photo for it ? Can I adjust the image size and mempercepan the webcam image capture process ? Thank you
reply : yusril.edmodo@Gmail.com
Hi Yusril,
CommandCam only captures images at the camera device’s default resolution. However, you can use a command line image processing tool like ImageMagick to automatically scale the captured image to a different size. For certain situations that may be an adequate solution. If you really need to specify the capture resolution, you can try another program I wrote called RobotEyez which provides more flexibility for specifying the resolution. Here’s the link:
https://batchloaf.wordpress.com/2011/11/27/ultra-simple-machine-vision-in-c-with-roboteyes/
Note that to save the image as a BMP file with RobotEyez, you’ll need to use the “/bmp” command line option.
Ted
hi, what would be easy if this tool will works for Win7 and a HD webcam. Any idea to fix that or improvement could be done to cover more webcam to with supporting devices.
A quick test shows no positive result but failed with error message below.
https://batchloaf.wordpress.com/CommandCam
This version 24-1-2013
Capture device: 1.3M HD WebCam
Error: 2147943850
Could not run filter graph
Hi Ted,
I tried this and it works great, Do you have anything similar in a dll, i have been playing with directshow for a while and just need an app to capture a frame from a camera every 15 seconds. When i use directshow the cpu usage is up to about 6-7% just for this small app. I think this is because it connects to a camera and stays connected, you program seems to connect, take a snapshot and then disconnect. I am not very good with C++ as i have only played with VB and VB.net is there a way to recompile your code into a dll where i send it the serial number and it returns a bitmap.
Cheers
Stu
Your program is very nice
I tried to recompile it using Visual C++ 2015 in IDE but I faced to many errors .. I fixed everything
Just that remain
I get error on this line :
HANDLE hf = CreateFileW(“image.bmp”, GENERIC_WRITE, FILE_SHARE_WRITE, NULL, CREATE_ALWAYS, 0, NULL);
if (hf == INVALID_HANDLE_VALUE)
i
t says
Severity Code Description Project File Line Suppression State
Error C2664 ‘HANDLE CreateFileW(LPCWSTR,DWORD,DWORD,LPSECURITY_ATTRIBUTES,DWORD,DWORD,HANDLE)’: cannot convert argument 1 from ‘const char [10]’ to ‘LPCWSTR’ abc d:\webcam\abc\source.cpp 585
I tried to modify the code make it compatible with VC2015 but I failed , I disabled Unicode but other part of code is not compatible to it ..
Please advise !
Thank you
batchloaf,
Thank you for the excellent program. I have it working on Windows 10 / 64-bit PC without any hiccups. Like other programmers I have a few questions & depending on the answers a wish list.
1. Can you specify where the webcam window pops up on the screen?
2. Rather than use /delay, is there a way to have a button or trigger be used for the snapshot capture?
Russ
I’m fascinated with both CommandCam and RobotEyez. I’m using this to monitor a machine. Works great. However initial camera delay when warming up is driving me crazy. RobotEyez solve this problem when using /frames -1 flag since next picture is taken immediately. Unfortunately I need to trigger next picture not in a fixed interval. I need to take picture when some process finish. I need to keep camera “up” so next picture is taken immediately. I tried using /period 1000 /command WaitForProcessFinish.exe that is a program that waits for a process to be finished. Then it closes and RobotEyez continue taking pictures every 1000 msec. So far so good. HOWEVER somehow next picture taken is not taken live. It’s grabbed from some buffer. I guess RobotEyez take picture every /period milisecs, no matter if /command execution takes longer to finish. Is there any way to overcome this? Thanks for your time. Francisco
Hi Francisco,
I think your frustration with the startup delay is shared by other users. Unfortunately, I don’t think much can be done to reduce the delay directly, since it seems to originate below application level (i.e. in the operating system).
A possible approach which I have been giving some thought to is to create a modified application that effectively runs as a kind of background service (let’s call this the “server”). It would keep the camera open continuously while it’s running. A second tiny command line app (let’s call it the “client”) would be able to trigger the server to grab a frame of the video and write it to a file. That way, the snapshot would happen almost immediately. It would take me some time to implement this, but I might give it a try.
Ted
Hi Ted. In the description of the client component, if you’d be willing to replace “write it to a file” with “write it to file or send it back to stdio” and you might really be onto something. 🙂
Hi Joshua,
Do you mean that when the user runs the client to request a snapshot, the server would capture a frame and output the raw pixel values to stdout?
Ted
Either that, or the pipe of the users choosing (I don’t know how that would work, though). I think that stdout might be the easiest way for the caller to consume the data as soon as it’s available.
Can this program be used for capturing images from DSLR cameras from Canon or Nikon?
Hi Ali,
I’ve never used it for anything like that. If the camera can be connected to the PC via USB and opened the way a webcam would be then it will probably work. However, if the computer only accesses the camera as a storage device (similar to a USB key) then it probably won’t work with Command Cam.
So basically, it depends on the camera and how it communicates with the PC.
Ted
I am having problems on my Microsoft Surface Pro 3 trying use the preview feature in CommandCam via commandline. It works fine otherwise, but if I try to add in the preview feature, it says “Could not render preview video stream.” I checked and both cameras on the Surface have 1080p video capture capability, which happens to be precisely the resolution at which CameraCam has been saving pictures….so that leads me to believe that resolution is not the issue here.
Do you have any suggestions or ideas? Anything you could do to help would be much appreciated!
For the heck of it, I uploaded the commandline history into pastebin so that you could see with your own eyes: http://pastebin.ca/3662875. I first ran it with the preview argument and it failed, and then I ran it without it and it succeeded.
Hmm, that’s strange. It’s a while since I’ve been looking at this DirectShow video capture stuff and even when I was writing CommandCam I found it very confusing, but as far as I recall the “capture filter” (which is the software object that represents the camera) can have two output “pins” – a capture pin and a preview pin. Maybe it’s possible that the drivers for the cameras on your device don’t support both pins for some reason?
Ted
Wow you are on top of things! Fastest response ever!
I’m not sure about the output pins or how I would even find that out. Do you have any suggestions? Or can you think of anything at all to test to try to see what the issue is?
Thanks by the way for your program. I spent a lot of time searching for something that would fulfill my purposes, and this is the only thing I found that would work. So far it’s all great except for the preview issue (which unfortunately may be somewhat of a dealbreaker if I can’t resolve it). I am hoping to use it in field data collection for scientific monitoring of habitat restoration 🙂
Whoops. Sorry, meant to reply to your response, not in a new comment.
Hi Timbo,
The “output pins” on the video capture filter aren’t physical pins at all – they’re basically just virtual signal outputs from the video capture driver for the camera. DirectShow is a Microsoft library for developing video and multimedia applications. In DirectShow terminology (of which there is a lot!) video processing applications are made by connecting together software modules called “filters” in a structure called a “graph”. A DirectShow filter might be a driver for a camera, or a video codec, or a software block that displays the video on the screen, or a software block that writes a video stream into a file. When a developer writes a DirectShow application, he/she connects a string of these filters together to form a complete system that does something with a video stream. Connections are formed between two filters by connecting an “output pin” on one filter to an “input pin” on another filter. Again, these are not physical pins – they’re all just software objects.
Anyway, as I understand it, when a camera manufacturer provides a DirectShow driver for their device, they are basically providing a “capture filter” with one or more output pins that can be connected to other DIrectShow filters. Many capture filters provide a preview output pin and a capture output pin. The exact difference is a bit hazy in my memory, but I think the crux of it is that the preview pin drops frames if it needs to in order to stay right up to date with the video stream, whereas a capture pin doesn’t drop frames. My suggestion that the capture filter provided by the driver for your camera might not include a preview pin is really just a guess, but either way I wouldn’t recommend trawling through any DirectShow documentation – it’s all absolutely mind-boggling.
A better approach would be to find an alternative program that does what you need. Have you tried VLC? It’s extremely flexible and rich in features and it has a command line mode of operation which can be used to automate video capture, etc. I would recommend giving that a try. You could start by googling “VLC command line video capture” and see what you find.
Ted
Oh, and VLC is free to download and install!
Ted
hi..can you pipe the output and send..for instance..to some tcp server?..thanks!
Hi rown,
I haven’t tried this myself, but I guess it’s possible. My instinct would be to try something like this:
1. Create a so-called named pipe.
2. Run CommandCam using the “/filename” option to specify the filename of the named pipe.
3. Use a program like netcat to connect the output of the named pipe to a specific port at an IP address of your choice.
Like I say though, I’ve never tried it!
Ted
Hi! I’ve updated CommandCam to support some common YUV image formats found in webcams (YUY2, I420) and MJPG. I’ve also added the possibility to select the size (resolution) of the captured image and the compression format. Detailed listing of capture devices is also improved – it now shows all supported compression formats, bits, resolutions and frames/sec for each device. Source code is available in my Github repo https://github.com/mrihtar/CommandCam The precompiled version with source code (CommandCam-2.6.zip) is available for download on https://app.box.com/s/gbeqxo1hpe3hm5jmv7l96oxje3vd0zbv
Thanks mrihtar, sounds like you’ve made some brilliant improvements! I’ll have to try it out once I’m back on a Windows machine. Based on some of the requests I’ve been receiving in this comment section, I’m sure a lot of people will find your update very useful.
Thanks,
Ted
Hi Ted, do you have any solution for ip camera??
Kamel
Hi Kamel,
No, I’m afraid not. CommandCam is really only designed to work with DirectShow video capture devices and I don’t think an IP camera would appear that way in Windows.
Perhaps you could try VLC? It’s free software and includes many flexible options for automatic image and video capture from a broad range of devices.
Ted
Many thanks Ted
Hi… I have downloaded the program few weeks ago.. it work as I really desire .. but I have a problem.. i want to ask… how to make the program save more than one photo … the one fore yesterday .. and the one for today… (how to save more than one photo at the same time)
Hi Siham,
To make CommandCam take more than one photo, you basically just run it multiple times. If you want to automate that process, a simple way is to write a batch (.bat) file which includes a loop and a time delay. You can find lots of tutorials about batch files online.
I don’t know if the Windows Task Scheduler still exists, but if it does that’s another option which might be appropriate if you want to run CommandCam once a day.
Ted
Sorry if this question was asked but could CommandCam take more pictures, so it does not delete last picture taken.
Hi Tony,
You could try something like this:
…or this…
I think that should give you a new file (named with the current time or date/time) every time you run CommandCam. Would that solve your problem?
%TIME% and %DATE% are built-in environment variables in Windows that should already contain the current time and date. I’m on a Linux machine here, so I haven’t been able to try to the above solution, but I think it (or something very like it) should work.
Ted
Hi!
first of all, Thanks for this great program, it has been useful. Recently appeared a problem whit it, when i run it, my web cam turns on but never take the picture and the web cam never turn off untill i disconnect it. Could you help me? i use it to record staff attendance
Hi Paulina,
Sorry, but I don’t know why that is happening. Did something change that could have caused it to stop working? For example, did you change the camera, the computer or the version of Windows?
If CommandCam won’t work for you anymore, I suggest trying ffmpeg. It can also be used to capture an image from the command line (or batch file) and it’s free software. It’s an active project too, so it will be updated more frequently than CommandCam.
Unless you want to compile ffmpeg from source yourself, you can download an executable version for Windows from here:
If you do a quick google, you’ll find examples of people using ffmpeg for image capture from USB webcams.
Ted
hello, is it that possible to capture and save image as TIF as standard fax will do, i.e. 1728 x 1145 pixels, and the black & white exactly like the facsimile received with A4 paper printed.
Hi Xiaolaba,
Sorry for the delay of several months in responding!
Short answer: no, CommandCam can’t do that. I suggest using ffmpeg to capture the image at the resolution you want and then using ImageMagick’s convert command to turn the image into a TIF format file. Both ffmpeg and ImageMagick and free / open source software. You’ll need to do a bit of googling to figure out the exact command line you need for each one, but I’m fairly sure they can do more or less what you’re trying to do.
Ted
Hi
first of all thanks for this program.
i wanted to know can i change the default capture device to my hp webcam
as it always take (AVerMedia devie) as the capture device as it is first available.
and one more thing i dont know anything about coding.
Try this:
i tried. but it only changes for that moment, after i open it again it reverts back to its original
settings.
and another thing can you help me make a batch file that will open this program
which is in another drive i.e d drive (if you are ok with it).
The /devnum setting only applies to that specific time you run CommandCam – it won’t chage which camera is the default one in Windows. To change the default camera, you’ll have to modify some Windows configuration stuff which is nothing to do with CommandCam.
To make a batch file, you just put the the CommandCam command that you want into a plain text file and save it as something.bat. For example, you could put the following in a file called “snapshot.bat”:
Ted
Thank you Ted with your help i made a batch file of this command
(CommandCam /devnum 2 /delay 5000 /filename %DATE%.bmp)
and kept that batch file in startup folder, so everytime i log in it will take a snapshot.
THANKS A LOT
Excellent! Glad to hear you got it working!
Ted
Hi Ted,
There is any way yo use an ip camera with your application?
Thanks.
Hi Kamel,
No, it can’t access an IP camera. I suggest trying ffmpeg or VLC. One or both of those can almost certainly grab an image (or video) from a remote camera over IP. They’re both free and will provide a lot more flexible options than CommandCam.
Ted
Thanks Ted
Pingback: Accessing WebCam with NodeJS - ExceptionsHub
Amazing!! I was making my own self serve and now you log in and it saves into a file!! Wow!! I would recommend this to anyone!
Also i had a plug and play cam, and it found the driver and worked, AMAZING!
And the different syntax allow much more custom commands!
Cool!
Ted
Hello,
I am trying to use this software to capture a photo of users attempting to log into my account.
so i used taskmanager and event viewer to get it done but so far no succes. should I place this executable in aspecific directory so windows hascontrol?
Hi Wim,
Sorry, I’m not sure how to do that. I think it’s possible, but I haven’t done it before and I don’t have Windows on my computer anymore, so I can’t test it.
Ted
On Windows 10, here is a way to make the program only take a snapshot on a failed login and not overwrite the image file.
First create a batch file (.bat) using a text editor like Notepad in the same folder as CommandCam with the following text:
@echo off
for /f “tokens=2 delims==” %%I in (‘wmic os get localdatetime /format:list’) do set datetime=%%I
set DNT=%datetime:~0,8%%datetime:~8,6%
timeout 3
rename image.bmp “%DNT%”.bmp
Then, save the batch file.
Go to the Event Viewer in C:\ProgramData\Microsoft\Windows\Start Menu\Programs\Administrative Tools
Open the Windows Logs>Security events list and search for an event with Event ID 4625. This is a failed login attempt. You can create one yourself by intentionally entering the wrong password on the lock screen. Once you locate an event 4625, select it, and on the right pane select Attach Task To This Event. Follow the steps and hit next. Under Action, select Start a program, hit next. Browse for CommandCam.exe, add argument /quiet without quotation marks, hit next and finish.
Now go to the Task Scheduler in C:\ProgramData\Microsoft\Windows\Start Menu\Programs\Administrative Tools
Open Task Scheduler Library>Event Viewer Tasks task list. Find your task using the name you gave it or the default name, right click on it and go to Properties.
On the General tab, select Run whether user is logged in or not, and deselect Do not store password. Select Run with highest privileges. Select configure for (in this case Windows 10 but use your operating system).
On the Actions tab, click New… then search for the CommandCam.bat file you made in the first step.
On the Settings tab, select Allow task to be run on demand and select Run a new instance in parallel from the drop down menu. Hit OK and type in your user password so the system can run the task from the Windows login screen.
And you’re done!
The files should be saved as timestamped bitmaps in the directory CommandCam is located.
Note that you might have to set the Start in folders to the CommandCam directory for both actions on the triggered task.
Thanks Rafael. Sorry, I just saw this now because I haven’t had time to keep on top of my blog comments. Otherwise I would have made this useful tip visible immediately!
Ted
Also, do not use quotation marks for the start in folder paths.
Hello. Is there any way to set up CC so that it automatically takes a picture when someone enters any PW to try to get my laptop out of sleep mode? Also is there any way to turn off the flash so whoever tries to enter a PW does not see it flash and realize it took a picture of them?
Hii, How to change the resolution of captured images?
Hi Ted. Thank you very much for this app. I have created an Access application that several local food pantries are using. On of the sites asked for the ability to have a picture of the client in the database. Your app was the only way I found to enable that functionality. But, there is one little thing that I found, I could not get it to work unless I used the /preview parameter, even when I launch CommandCam at the DOS prompt.
Just found this and wanted to help out as I had the same issue as a few people here….
I wanted to run a photo capture on startup… to do this follow the procedure here (with modifications below).
https://www.groovypost.com/howto/snap-picture-log-in-save-to-dropbox/
*Note: to find the startup folder on Windows 10 – go to Start, Run: shell:startup
but instead also create a .bat file (google it) and add the code below and save it.
Then use the whatever.bat to run on startup etc, this will run the CommandCam.exe, rename the standard filename to the local folder and then delete the original.
@echo off
CommandCam.exe
for /F “tokens=2” %%i in (‘date /t’) do set mydate=%%i
set mydate=%mydate:/=-%
set mytime=%time::=-%
set filename=%mydate%-%mytime%.bmp
copy image.bmp %filename%
del image.bmp
—
What I am looking to do now is run the same procedure on Login, and will test using http://www.eventghost.net to see if I can manage this – Ill update what I find.
this, instead:
@echo off
CommandCam.exe
for /F “tokens=2” %%i in (‘date /t’) do set mydate=%%i
set mydate=%mydate:/=-%
set mytime=%time::=-%
set filename=%mydate%-%mytime%.bmp
ren image.bmp %filename%
Ive made this work on login using Event Ghost – I’ll post an update on my site if anyone is interested, may require video and screen shots.
Thanks Ted BTW – useful tool.
https://stuartread.com/eventghost-login-image-capture
Note that you might have to set the Start in folders to the CommandCam directory for both actions on the triggered task.
Also, do not use quotation marks for the start in folder paths.
I’m using this for when I or someone else logs onto my pc but its only keeping one photo. Is there a way to have it time & date stamp each photo so they’re individual?
As much as I’m decent with a pc, can you break it down for an idiot?
Pingback: SerialSend – Hallo, here is Pane!
Hi,
Thanks for the great app. I am using it for a long time and wondering if you are going to implement changing quality or resolution functionality. It would be great since sometimes each capture takes quite amount of space.
Looking forward to your update. AFTER 8 years..
Regards,
Semih
Thank you for all the work you have done on this. I use it in an accde program to capture different images. I have the preview available for 3 to 5 seconds. The only issue I have to date is that The preview box opens in different areas depending on the screen and pc the access db is run on. There is no way to set the location that I can see. Is it possible to open the preview in fullscreen?
Hi,
this tool worked great for a few days (ran in a loop), then suddenly one day all new images were just showing the color purple. I stopped the loop and ran the command “CommandCam /devlist”, and it returned “No video devices found”.
Any idea what might have happened here?
Thanks,
Matts
Hi Matt. It sounds like the problem might be outside of CommandCam – like a driver issue or something like that. Did you try opening the camera with a different program, like VLC or Skype, to see if the cam is working elsewhere? Ted
Hi Dr. Burke,
Thank you so much for working on this! I am trying to connect it to a Canon Rebel T3i. I downloaded the EOS Webcam Utility so that the camera can be used as a webcam (works well with Skype, OBS, etc.) The device is recognized in CommandCam /devlist. When I use this device, it takes a photo (I can hear it) and says:
Capture device: EOS Webcam Utility Beta
leaving aero onleaving aero onCapture Resolution: 1280×750
Captured image to image.bmp
Then when I open the .bmp file that it created, it is unable to open/convert/etc.
I’ve looked into why it might be saying “leaving aero on” and it might have something to do with the aero theme from Windows 7. But I am running this on Windows 10 so I am a bit confused.
Any advice you have would be greatly appreciated!
Pingback: ¦› Seguridad informática: Tome una foto de quienquiera que se conecte en su portátil
BUENAS noches descargue el commandcam y lo adecue a mi proyecto necesito que el comandcam pueda tomar varias fotos y guardar esas fotos en una misma celda como se haria
Pingback: USB kamera képkészítés parancssorból, időzítve Windows 10-en (timelapse parancssorból) – Geeklány
Hey Batchloaf,
We use this tool to help make badges for new employees. It has been a savior! So cool!
We recently had it “go down”.
Overall, the message that populates when I type “CommanCam” in the command prompt is spot on. On your screenshot within the directions provided above, it states:
Capture device: USB Video Device
Capture resolution: 640X480
Captured image to image.bmp
When I run the CommandCam in the command prompt, my results:
Capture device: USB Camera
and then it sits.
I’ve taken pictures with the camera, so the camera itself works. Any tips/tricks on how to push this through?
Thanks a million,
Kim
Hi Kim,
Unfortunately, I don’t have a Windows machine any more, so I haven’t been able to properly maintain CommandCam. Hence, I can’t test to see if I get the same problem you’re having. Did you switch to a different camera or anything that might explain the change in behaviour?
Since I’m not in a position to investigate the problem properly, my best advice is to try using a different program. An ideal candidate would be ffmpeg, which is free software, mature and very widely used, so it’s well maintained. It will probably also allow you to capture much better resolution images with today’s high resolution cameras. ffmpeg is a little bit of a pain to install on Windows, but you’ll find instructions online. Once you have it installed and you figure out the right command, you’ll be able to use it the same way you’ve been using CommandCam, but it will probably be a more future-proof solution and will provide better quality images.
Ted
Ted –
Wonderful! Thank you so much Ted! If you can figure out the right command in ffmpeg, you would be a life-saver—again! We had lightening strike the mill that I work in a week ago, and since then…it’s been one thing after another. Any help you can provide in order for me to hit the “easy” button would be greatly appreciated!
I’ll keep an eye out for updates.
Glad I mentioned it – didn’t realize it would be a *global* problem.
Best,
Kim
Heres some nodejs code that’ll use ffmpeg to grab a snapshot:
module.takeSnapshot = function() {
var cmd = [‘-y’, ‘-f’, ‘dshow’, ‘-i’, `video=${video_device}`, ‘-vframes’, ‘1’, ‘c:\path\to\screenshot\webcam_snapshot.jpg’];
child_process.execFile(‘path\to\ffmpeg.exe’, cmd, function(err, stdout, stderr) {});
return;
}
Hi Kim and Joshua,
I did a little bit of testing and it’s very easy to use ffmpeg to capture a still image from a USB camera. On Linux, I was able to get it working in 30 seconds. The main challenge on Windows is actually just getting ffmpeg installed because they don’t provide the usual kind of installer. Below, I’ve set out the process that I think will work on Windows, but since I don’t have Windows on my own machine I haven’t been able to test it. If either of you try it and can confirm that it works then I’d appreciate you letting me know so that I can add some information about it at the top of the page. Anyway, here’s the process…
Go to the ffmpeg download page:
https://ffmpeg.org/download.html
Click on the Windows logo to jump to the Windows EXE files section. There seem to be two different Windows builds – one by gyan.dev and another by BtbN – you’ll see links to both there. I have no idea which one is preferred – probably both will work. Either way, you’ll have to download some kind of zip file and extract it to get the actual program.
There are various instructions online for installing ffmpeg on Windows. Just google “ffmpeg windows install” and take your pick. Since I don’t have Windows on my machine, I haven’t been able to verify any particular set of instructions, but the ones I looked at were all broadly similar.
Anyway, once you have ffmpeg installed, try the following command to list available DirectShow video capture devices:
Hopefully you’ll see your USB camera listed under a recognisable name. To capture a still image from the camera, use the following command (replace “Integrated Camera” with the name of your camera device, exactly as it was shown when you listed the available DirectShow devices):
It’s possible that the camera might take a little time to adjust to the light level after capture begins, so if you want to capture multiple frames and then use the last one, you can try the following command (this example captures two frames as numbered image files, but you can change the number to capture as many frames as you like).
You’ll find lots of example commands online for changing the image resolution and other settings.
Ted
My exe’s of CommandCam just disappeared. As in, just today. One minute it’s there, one minute it’s not. I had dozens of copies of the the exe, and over the course of a few minutes they all disappeared. Was this a programmed autodestruction, or was the file flagged as Malicious Software?
Latest update of Microsoft antivirus defender detects it as a threat, and remove it. Just found out the same problem You can recover this file back from Windows Defender, but now i am not sure what to do. I was using this file for years…
Here is screenshot how Windows Defender detects threat in this file: http://sergeskor.no-ip.org/Files/Commandcam_screenshot.png
Hi Joshua and Serge,
Hmm, this is an annoying problem. I’ve no idea why Windows Defender would suddenly have decided CommandCam is a threat. There’s nothing in there other than what you see in the source code above. I suppose you could try compiling the source code yourself? Maybe the resulting exe file would be exempt from detection then?
If that doesn’t work, or you don’t want to get into messing with compiling code, then I suggest trying ffmpeg as an alternative to CommandCam. It’s a little bit tricky to install on Windows, but you’ll find instructions online. Once you have it installed though, and you figure out the right command line, you’ll be able to use it the exact same way you’ve been using CommandCam. Even better, it will support higher resolution capture than CommandCam and can do a lot more than still image capture too. It’s free software and very widely used, so it’s tried and tested. I use it all the time in Linux and it’s an absolute gem of a program.
Presumably this Windows Defender issue is going to affect other people who are using CommandCam too. I’ll see if I can figure out the right command line to use ffmpeg as a drop-in replacement. If so, I’ll add something at the top of the page.
Ted
Ted,
thank you for your comments. Just for your information – Symantec antivirus does not detect any “threats” in CommandCam.exe yet :-). Only Windows 10 Windows Defender, updated yesterday or today, found “threat’.
Pingback: Computer Security: Snap a Picture of Whoever Logs on Your Laptop – Tech Blog
Pingback: #StackBounty: #windows-10 #command-line #freeze #threads Console program randomly freezes – TechUtils.in
How do i take multiple pictures with this
Hi NmNa. You would need to run it multiple times. You could set up a batch file with a loop to run it repeatedly.
The CommandCam.exe file on Google Docs can’t pass the virus scan. Tried twice. It got deleted. The one on github (the 2012 version) is fine.
Hello @batchloaf,
Good job on this tools ! I’m actually coupling it with a personnal software I’m using for security issues. It took a picture each time someone use or try to use a computer after a short period of inactivity. I wonder if there is a way to activate the camera in silent mode ? I mean, with the webcam LED off.
Hi JP. That won’t be possible on most cases because (for privacy protection reasons) the operating system won’t allow the camera to operate without the indicator light being on, or else the light is actually hard wired to the camera power so that it’s physically impossible for the computer to activate the camera without also switching on the indicator light.
Ted
Pingback: Win7Pro: Capturing a photo and recording voice when someone failed to login - Boot Panic
Pingback: How to display video streaming from USB cam using DirectShow? - TechTalk7
Thanks for sharing informative article.
Hello, great program, love it.
I have made a batch file that creates a image with a incremented value (for ex: image.bmp, then image1.bmp, then image2.bmp etc…) with this program.
Anyone who wants to use this can use it.
Batch File Code:
@echo off
setlocal
set imgnum=
if not exist image.bmp goto nonexisting
set imgnum=0
:existing
set /A imgnum += 1
if exist image%imgnum%.bmp goto existing
:nonexisting
.\CommandCam.exe /filename image%imgnum%.bmp
endlocal
@batchloaf great program I love it, now I can add this to my startup program list and make a photo time lapse lol 😀
Wow, that’s very useful! Thanks Unknown Dude!
The software work fine, but is detected as malware by Windows 10 Defender Antivus. I add a exception in antivirus software for this.
Thanks Ramón. Others have had the same issue. I’m not sure why Windows antivirus is doing that, but I’m glad you were able to get it working anyway.
Ted