Documentation for C-Werk 2.0.

Previous page Camera and image requirements for License plate recognition (VT)  Fine-tuning the VT number plate detection tool Next page

Attention!

The license plate recognition requires a CPU that supports the SSE4.1/SSE4.2 instructions.


To configure ANPR, do as follows:

  1. Download ANPR (VT) from the website and install it. 
  2. Create the License plate recognition (VT) object (1) and select it.

  3. If you need to use this detection tool for real-time license plate recognition, set the corresponding parameter to Yes (2) (see Configuring real-time vehicle license plate recognition).

  4. If you need to enable recording of metadata to the database, select Yes from the Record objects tracking list (3).
  5. If the camera supports multistreaming, select the stream for which detection is needed. Selecting a low-quality video stream allows reducing the load on the Server (4).

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

    1. By LP coordinates—if the license plate coordinates shift from top to bottom, then the vehicle is moving towards the camera; if the license plate coordinates shift from bottom to top, then the vehicle is moving away from the camera;

    2. By LP scale—if the license plate scale increases, then the vehicle is moving towards the camera; if it decreases, then the vehicle is moving away from the camera.

  7. Select detection tool working mode: standard or neural (2). 

  8. In the appropriate fields, select one or more countries for ANPR (3).

    Attention!

    • The more countries you select, the slower the recognition works and the greater the likelihood of the recognition error.
    • To recognize the old USSR license plates, you need to select the corresponding country in the list.
  9. Select a processing resource for decoding video streams (4) (see Hardware requirements for GPU based video decoding). 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.

  10. Analyzed framed are scaled down to a specified resolution (5, 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 (down to 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.


  11. Select the current license type (6).

    LicenseDescription
    Search in archive

    Basic ANPR Search License.

    Attention! This type of license provides a 30-sec delay between the number recognition and the corresponding event (see Vehicle number plate recognition and search).

    Standard (25 FPS or 6 FPS)

    		The 25 FPS license has a priority. The 6 FPS license is used only if you don't have 25 FPS license. You cannot use the license plate recognition function if you have not purchased both licenses.

    Fast speed (25 FPS)

    		The "Fast" license allows you to process video feeds up to 25 FPS with a maximum vehicle speed of 150 km/h. You cannot use the license plate recognition function if you have not purchased the license.

    Slow speed (6 FPS)

    		The "Slow" license allows you to process video feeds up to 6 FPS with a maximum vehicle speed of 20 km/h. You cannot use the license plate recognition function if you have not purchased the license.

    Attention!

  12. Set the maximum and minimum width of the vehicle license plate as a percentage of the FOV width (7, 8).

    Attention!

    The Minimum LP width parameter affects the CPU load as follows: the higher the parameter is, the higher the load.

  13. Choose a processor to run detection on (9, see Hardware requirements for License plate recognition (VT), General information on configuring detection). In standard mode, detection runs on the CPU only. In neural network analysis mode, the algorithm runs on the NVIDIA GPU.

    Attention!

    Starting with Detector Pack 3.11, Intel HDDL and Intel NCS aren’t supported.

  14. Set the number of frames required for LPR/ANPR (10). This is a necessary condition, but it is not sufficient for the first output. This condition delays the LPR output. This parameter allows increasing the reliability of the results, as well as hiding false positives.

  15. If necessary, configure the advanced detection settings (see Fine-tuning the VT number plate detection tool).

  16. You can configure the ANPR area in 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.
  17. Click the Apply button.

Configuration of ANPR is now complete.

  • No labels