Axxon Next

FindFace Multi integration with Axxon Next allows you to detect and identify faces in video streams from an Axxon-based security system.

Important

One FindFace Multi instance supports interaction with only one Axxon Next server.

Integration with Axxon Next is implemented via the ffsecurity_axxon plugin.

To configure the FindFace Multi integration with Axxon Next in Ubuntu, do the following:

  1. Activate the plugin by uncommenting the INSTALLED_APPS.append(‘ffsecurity_axxon’) line in the /etc/findface-security/config.py configuration file.

    sudo vi /etc/findface-security/config.py
    
    ...
    
    # ==============================================================================
    # FINDFACE SECURITY PLUGINS
    # ==============================================================================
    # Uncomment lines below to enable plugins. Please consult documentation for
    # a plugin specific settings.
    # =============== Axxon ================
    INSTALLED_APPS.append('ffsecurity_axxon')
    
  2. Uncomment the FFSECURITY->AXXON section in the configuration file. Fill it out as shown in the example below. In the api parameter, specify the IP address of the Axxon Next server that will provide FindFace Multi with Axxon API and HLS-archive streams. In the rtsp parameter, specify the common segment of Axxon video stream addresses. name, user, password: the Axxon Next server name and credentials to access it.

    FFSECURITY['AXXON'] = [
        {
            'name': 'server_name',
            'api': 'http://example.com/',
            'rtsp': 'rtsp://example.com:554/',
            'user': 'user',
            'password': 'password',
        }
    ]
    
  3. (Optional). If facial recognition events are required to contain video from Axxon Next, uncomment the FFSECURITY_UI_CONFIG['dossier'] section.

    FFSECURITY_UI_CONFIG['dossier'] = {
       'video': True,
    }
    
  4. Create representations of Axxon Next cameras in FindFace Multi (see Camera Management). A camera representation URL must be specified in the format axxon:<friendlyNameLong>, where friendlyNameLong is a camera name on the Axxon Next server. Find out this name in the Axxon user interface, or via Axxon API by executing:

    curl http://user:password@127.0.0.1/video-origins/
    
    {
       "OLOLOE-DEV/DeviceIpint.vhod_1/SourceEndpoint.video:0:0" : {
          "friendlyNameLong" : "vhod_1.Vhod_1",
          "friendlyNameShort" : "Vhod_1",
          "origin" : "OLOLOE-DEV/DeviceIpint.vhod_1/SourceEndpoint.video:0:0",
          "state" : "signal_restored"
       }
    }
    

    For the camera from the example above, URL must be specified as axxon:vhod_1.Vhod_1.

The configuration is now finished. If the integration is properly configured, FindFace Multi will be detecting and identifying faces in Axxon Next video streams, and facial recognition events will be featuring video clips from Axxon Next (upon relevant settings).