Documentation for Face Intellect 7.3. Documentation for other versions of Intellect is available too.

Previous page Face Intellect configuration and setup procedure  Configuring the Face Recognition Server object Next page

Face Detection software module functionality

The Face detection object is created on the basis of the Camera object in the Hardware tab of System settings dialog box.

Configure the Face detection object as follows:

  1. Go to settings panel of the Face detection object.

  2. Set the face search area border on the video image by specifying the required values in percent in the corresponding fields (1), or by clicking the  button (2) and using the interactive interface.

    Attention!

    Setting the face search area border is avaiable only if the Tevian face recognition module is used.

    After you click the  button, the window with the corresponding video camera image will open.

    To set the face search area border, click the Stop video button at the required time, select the required area using the left mouse button, and then click OK.



    The specified face search area border values will be automatically entered in the Left, Right, Upper and Lower fields of the Search area border % group.

  3. There are several variations of the face detection operation depending on the setting of the external capture and Face Recognition Server (3):

    1. External capture is NOT selected and the detection is NOT added to any of the Face Recognition Server channels. In this case, only capturing of photos of faces detected by Intellect software capturing is performed, and generation of vectors is not performed. This option is not for Face-Intellect.
    2. External capture is NOT selected but the detection is added to any of the  the Face Recognition Server channels. In this case, capturing and saving of photos of faces detected by Intellect software capturing, generation of vectors and recording them to database are performed. This option does not allow face recognition (while search and visitor counting are available).
    3. External capture is selected but the detection is NOT added to any of the Face Recognition Server channels. In this case, capturing and saving of photos of faces detected by Cognitec 8.8 or Tevian 1.8.1 capturing and generation of vectors are performed, but the vectors are not recorded to the database. This option is only suitable for use in scripts.

    4. External capture is selected and the detection is added to any of the Face Recognition Server channels. In this case, capturing and saving of photos of faces detected by Cognitec 8.8 or Tevian 1.8.1 capturing, generation of vectors and recording them to the database are performed.

      Important!

      It is required to enable 64-bit modules if the Tevian module is used for external capture (see Administrator's Guide, the Configuring of using 64-bits modules section).

    The external capture is recommended for use. From the External capture drop-down list select the Face Recognition Server with the help of which the face capturing with the subsequent generation of a vector (if the Create FIR in detection unit checkbox is set on the settings panel of the corresponding Face Recognition Server object – see Configuring the Face Recognition Server object) and saving of photos will be performed.

    Note.

    Proper configuring of the face recognition server selected as external capture is performed with the help of the Cfgedit.exe utility (detailed information about this utility see in the Cfgedit.exe utility for configuring the Cognitec face recognizer section).

  4. The Maximum performance mode is enabled by default (4). This mode allows improving the quality of face capture and recognition by using all available processor cores and increasing the RAM consumption. When this mode is disabled, the memory consumption will decrease, but faces can be missed at capturing.
  5. To pixelate the part of video image in which the face was recognized, do the following:
    1. To enable the pixelating function, set the Pixelate area checkbox (5).

    2. Set the Grain size cursor in position corresponding to the pixelation grain size in pixels (6).

  6. In the Frame rate, fps field (7), specify how many fps from the video stream are processed by the Face detection program module. This parameter is necessary for reducing the load on the Server.
  7. Specify the size of the captured face in percent relatively to the sizes of camera’s video surveillance window if the captured face is required to be dynamically framed (8).
  8. If the color of a dynamic frame needs be changed, left-click the Color field (9). Select the required color in the standard Color dialog box and click OK.

  9. If it is necessary to convert the color video image to black and white image to generate the vectors by the black-and-white frames, then uncheck the Process color video checkbox (10).

    Important!

    Generation of vectors by black-and-white frames is supported by Tevian 2.3.0 face recognition module only. It reduces the CPU load, insignificantly affecting the face capture and recognition quality.

  10. By default the Face detection tool processes a video with the resolution set for the Camera parent object. To process a video with 320x240 pixels resolution, set the Process video in low resolution checkbox (11).

    Note.

    Enabling this option reduces the load of the Server’s processor, but the probability of face capture also reduces.

    Attention!

    The Process video in low resolution setting is not relevant in case, when a high resolution is set for a Camera parent object. Also this settings is not in use if the Tevian module is applied for face capture.

  11. Set the minimum and maximum width and height of the face on the video image in pixels in the corresponding fields of the Face size, pixels group (12).

    Note.

    Faces will not be captured if the specified minimal sizes highly differ from actual sizes. For example, the detector will not capture big faces if it is configured for capturing small faces, and vice versa.

    Note.

    If all face sizes are set to 0, the Face Detection tool uses similar parameters set on the Face Recognition Server object settings panel (see Configuring the captured face size for recognition).

  12. Set the Zoom slider to the position corresponding to the required scaling of a captured face in the frame for recording (13). The current scale value is displayed in the field above the slider and ranges from 1 (only the captured face is recorded) to 5 (the whole frame is recorded).

    Note.

    This setting is relevant in case the Face Intellect system is installed. A frame with the captured face is recorded to the < Face-Intellect>\ Bmp\ person\ folder.

  13. To save the changes click the Apply button (14).

Note

After adding the Face Recognition Server object (see Configuring the Face Recognition Server object), in the field (6) of the Face Recognition Server debug window (see Debug window interface) the video resolution and the video stream number will be displayed.

Configuring the Face detection object is completed.