Documentation for C-Werk 2.0.

Previous page Examples of correct and incorrect video images  License plate recognition (RR) Next page

To configure License plate recognition (IV), do as follows:

  1. Submit to the technical support the MAC address of the Server where the detection tool will be used.

    Attention!

    The detection tool will not operate on the Server with a different MAC address.

  2. Copy the license received from technical support into the C:\Program Files\Common Files\Grundig\DetectorPack\LicenseFile_LprIV.txt file.

  3. Restart the Server (see Shutting down a Server, Starting a Server).

  4. Create the License plate recognition (IV) object and select it.

  5. If you need to use this detection tool for real-time license plate recognition (comparison with the license plates lists), set the corresponding parameter to Yes (1, see Configuring vehicle license plate recognition based on the created lists).

  6. If you need to record metadata, select Yes from the Record objects tracking list (2).
  7. If the camera supports multistreaming, select the stream for which detection is needed (3). Selecting a low-quality video stream allows reducing the load on the Server.

  8. Select the Algorithm for detecting vehicle direction (1):

    1. By LP coordinates: if LP coordinates change position from top to bottom, the vehicle moves towards the camera. If LP coordinates change position from bottom to top, the vehicle moves away from the camera.
    2. By LP scale change: if LP scale increases, the vehicle moves towards the camera. If LP scale decreases, the vehicle moves away from the camera. 

  9. Select the country from the list for LPR and the level of recognition accuracy (2).

    1. High load, high recognition accuracy—provides the maximum recognition accuracy, but creates a high load on the CPU and/or GPU.

    2. Medium load, medium recognition accuracy—provides a high recognition accuracy, requires less computing resources than the maximum accuracy.

    3. Low load, low recognition accuracy—provides the fastest recognition speed, but the recognition accuracy is low.

  10. Select a processing resource for decoding video streams (3). When you select a GPU, a stand-alone graphics card takes priority (when decoding with NVIDIA NVDEC chips). If there is no appropriate GPU, the decoding will use the Intel Quick Sync Video technology. Otherwise, CPU resources will be used for decoding.

  11. By default, the analyzed framed are scaled down to a specified resolution (4, 1920 pixels on the longer side). This is how it works:

    1. If the longer side of the source image exceeds the value specified in the Frame size change field, it is divided by two.

    2. If the resulting resolution falls below the specified value, it is used further.

    3. If the resulting resolution still exceeds the specified limit, it is divided by two, etc.

      Note

      For example, the source image resolution is 2048*1536, and the specified value is set to 1000.

      In this case, the source resolution will be halved two times (512*384), as after the first division, the number of pixels on the longer side exceeds the limit (1024 > 1000).

      If detection is performed on a higher resolution stream and detection errors occur, it is recommended to reduce the compression.

  12. Set the number of frames processed per second by the detection tool (5). The value should be in the range [0,016; 100].
  13. Set the maximum number of processor cores available for the detection tool. The 0 value means that all cores are used (6). The value should be in the range [−1; 1].
  14. Set the maximum and minimum width of the license plate as a percentage of the frame width (7, 8). The value should be in the range [1; 100].
  15. Set minimum quality of LPR (9). The higher the minimum recognition quality, the fewer false positives will be detected. The value should be in the range [0; 100].

  16. Select the processor for the detection tool—the CPU or one of NVIDIA GPUs (10, see General information on configuring detection).

    Attention!

    It may take several minutes to launch the algorithm on NVIDIA GPU. 

    If there are several GPUs in the system, a specific NVIDIA GPU value can be assigned to each License plate recognition (IV) tool.

  17. In the Event timeout field (1) specify the time interval in seconds between the initial LP recognition and event registration. The value should be in the range [0; 3600]. The 0 value sets the event registration to the moment when the track disappears from FOV.
  18. Specify the maximum and minimum number of characters in LP (2, 3). The value should be in the range [1; 20].

  19. In the Tracker timeout field (4), enter a time period in seconds after which the vehicle track is considered lost. The value should be in the range [0; 3600].

    Note

    This parameter should be used to prevent double detections that occur, for example, when the LP is recognized in the frame then gets hidden behind a visual obstacle and after that gets recognized by the detection tool again. If you set the timeout longer than the time of possible LP overlapping in the frame, the detection tool will trigger only once. 

  20. You can configure the LPR area in the FOV. The area is resized by moving the anchor points . 

    Note

     For your convenience, you can click the button and configure the mask on a still frame. To undo, click this button again.

    Detection area is displayed by default. You can click the button to hide the area. To undo, click this button again.

  21. Click the Apply button.

Configuration of License plate recognition (IV) is now complete.

  • No labels