To receive video and audio from USB webcams via the FFMPEG driver, you can add the Video capture device object either by Creating and configuring the Video Capture Device object or using the Camera discovery tool. The latter method is preferable.
Using the FFMPEG driver for connecting USB webcams has the following features:
- Video and audio streams are not combined for the discovered devices. For example, if there is a USB camera with a microphone, two devices are discovered—one video-only device and one audio-only device.
- In order to make a single device with both audio and video streams, manually edit the IP address field on the settings panel of the Video capture device object. To specify the audio device, enter its name with a colon after the name of the video device. See also Advanced settings for the FFMPEG device below.
- Both video and audio devices can be unavailable. For example, if you specify the device address as dshow://:Internal Microphone (Conexant ISST Audio), a device without video but with audio is available.
Creating the Video capture device object
To add the Video capture device object to receive video and audio via RTSP and RTMP protocols using the Dshow device (USB camera) model, do the following:
- Start the Camera discovery tool.

- In the Model list (3), select Dshow device (USB camera).
- In the Firmware list (4), select auto.
As a result, the discovered devices are displayed in the FFMPEG list (2). If the required device isn't discovered, proceed to the next step. If the device is discovered, proceed to step 5.
- Click the Search button (1) and wait for the repeat discovery to complete.
If necessary, you can add the device manually by Creating and configuring the Video Capture Device object. For the device address format, see Advanced settings for the FFMPEG device below. - In the Settings list (5), select:
- Video server settings, if you want to send the settings specified in Axxon PSIM to the camera when you add it.
- Device settings, if you want to use the device settings.
- Enter login (6) and password (7) to connect the FFMPEG device. The login and password are specified in the documentation of the network device that you connect. Usually these fields must be left blank.
- Check if the TCP/IP port (8) is correct. If necessary, enter the required value in the field.
- Select the objects that you want to create on the basis of the FFMPEG device by setting the corresponding checkboxes (9).
- Click the Apply button (10) for one device or the Apply button (11) if several devices are selected.
The selected objects are automatically created in the object tree on the Hardware tab. - Click the Exit button (12) to close the Camera discovery tool window.
Advanced settings for the FFMPEG device
- You can specify the address manually by including the parameters of the ffmpeg utility after the colon in the IP address field on the settings panel of the Video capture device object. In this case the address looks like this:
dshow[index]://[video_device_name]:[audio_device_name]:[-additional_options]
- By default, the index is set to 0 unless you specify another option. A non-zero index is used when several devices with the same name are connected to the system at the same time, for example, dshow1://USB 2.0 HD Camera if there is already dshow://USB 2.0 HD Camera.
- You can specify the required resolution and codec that the USB camera outputs different from the default settings:
-pixel_format yuv420p -video_size 1280x960
- The list of settings and the list of parameters of these settings depend on the format and a specific device. You can find out the set of allowed parameters for dshow devices using the following command:
ffmpeg -list_options true -f dshow -i video="<device name>"
Connecting multistream devices
A camera that is created on the basis of the Video capture device object with the Dshow device (USB camera) model has two streams. You can select a codec for each stream:
- YUV422—codec for the detector stream.
- MJPEG—codec for transmission over the network and saving to an archive.