Page History

To set configure the recognition parameters for the of captured faces recognition, do the following:

1. Go to the settings panel of the Face recognition server object, to the General settings tab (1) of the Face recognition server settings panel.
   Image Modified
2. Set the Select the best frame for recognition checkbox (2) if  checkbox to select the best frame from all images of one person which that came from the Face detection tool to the Face recognition server from Face detection it is required to select the best frame and create FIR only for it.
3. In the Similarity level field (3), enter specify the similarity level of between the captured face and the reference image that will be considered a recognized match. The Similarity level parameter value is specified in percentas a percentage, above which the face is considered recognized.
   

   Note
   title Attention!
   If you change the value, note that the new value applies to new faces. The similarity level for other faces remains the same.

4. In the Search for similar faces field (4), enter specify the maximum number of matching similar reference images, which identifiers and similarity levels level will be stored in database for one frame of a captured face.

   Info
   title Note
   It is recommended to specify We recommend specifying the 1 value for this field for a more precise search of the recognized faces (see see Filtering the recognized and unrecognized faces).

5. In the Minimum quality for face recognition (0-100) field (5), enter field, specify as a percentage the minimum quality of the captured face in percent, at which the recognition should be performedit is recognized. The default value is 60% 60%.
   

   Note
   title ImportantAttention!
   If Tevian recognition module is in use, it is recommended to set the Minimum quality for face recognition (0-100) value to 60% when updating from Face PSIM 7.0 to Face PSIM of later version.If face concealment recognition is enabled, then in order to correctly detect the presence of a face mask, it may can be necessary to set the  the Minimum quality for face recognition (0-100) value below the default value (see Configuring the permissions and additional settings).

6. In the Minimum quality for adding face to DB (0-100) field (6), enter , specify as a percentage the minimum quality percent for of face to be added to the database. 

7. Set the Create FIR in detection unit checkbox, so that the face detection tool generates and attaches biometric parameters to the face image. If the checkbox is clear, the detection tool only captures faces, and the Face recognition server generates and attaches the biometric parameters. By default, the face detectors for which the configured Face recognition server is selected as an external capture, generate and attach biometric face parameters to the image. Such procedure is performed if the Create FIR in detection unit checkbox is set (7). If it is necessary that the face detectors only capture the face, and the biometric face parameters are generated by the Face recognition server, it is necessary to deselect the Create FIR in detection unit checkbox.
8. Set the Select the best frame in detection unit checkbox (8) if it is required to select the best frames from all images received by Face detectors, for which the configured Face recognition server is selected as an external capture, and send only them , so that only the best frame is sent to the Face recognition server. If the checkbox is clear, all frames are sent to the Face recognition server.
9. See

   If it is required to skip repeated recognitions, in In the Skip repeated recognitions, s field (9), set field, specify the time period in seconds more than 0 during which a captured face wonisn't be recognized repeatedly. For details, see . The default value is 2. See also the IgnoreSamePeople registry key .

   Info

   title	Note

   in the Registry keys reference guide, for details about working with the registry, see Working with Windows OS registry.

10. If it is necessary for the recognition to be performed on a discrete Set the Use GPU checkbox to select the discrete graphics card when using the Tevian VA and VideoIntellect 1.1 face recognition modules, set . If the Use GPU checkbox (10). Otherwisecheckbox is clear, the recognition will be is performed on the CPU. 

    Note
    title Attention!
    If the Use GPU checkbox is set for the first time, then at the same moment the caching process will start starts in a new window to optimize the recognition on the current GPU. Do not interrupt this process. If for some reason the caching process was interrupted, it is necessary to delete the folder with the name like "GPU-[video card UUID]" at cache folder at <Face PSIM installation directory>\ Modules64\Firserver\FaceRecognition\TevianVa_ [SDK version]\ (for the VA module). Afterwards, when the Use GPU checkbox is set, the recognition using the GPU will be is performed immediately without the caching process. If you replace the video card, the caching process will restart restarts automatically. After the caching process is complete, enable the recognition using the GPU by selecting the video card from the Select GPU drop-down list in the settings of the Face Detection object (see Basic settings of the Face Detection module).

11. In the Collected no less than ... frames with person field (12) enter the minimal field, specify the number of frames to be collected o one person's face to start the face recognitionafter which a face is recognized.

12. In the Person in field of view for no less than … seconds field (13) enter field , specify the time in seconds upon reaching which the recognition will be performedstarts.

    title

    Info
    title Note
    The Person left the camera field of view option condition is always enabled (11) .
    Info
    	Note	The face recognition starts if at least one of the three conditions (11-13) is met from the Recognize if any of the conditions met settings group.

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

Setting Configuring the recognition parameters for of captured faces recognition is complete.