Page History

...

Starting with DetectorPack 3.10, a utility was added to the RR LPR add-on (see Installing DetectorPack add-ons), which allows you to create GPU neural network caches without using Axxon One. The presence of the cache speeds up the initialization and optimizes video memory consumption.

To optimize To optimize the operation of the the License plate recognition RR and Vehicle recognition RR on RR—Parking detector on GPU, do the following:

1. Stop the server (see Stopping the server).
   

   Note
   title Attention!
   If the system has the software running on GPU, it is necessary to stop its operation.

2. Login as ngp superuser:

   1. In the command prompt, run the command:

      Code Block
      language bash
      sudo su ngp

   2. Enter the password for the root superuser.

3. Create a folder with a custom name to store the cache. For example:

   Code Block
   language bash
   mkdir /opt/AxxonSoft/AxxonOne/gpucache

4. Change folder permissions:

   Code Block
   language bash
   chmod -R 777 /opt/AxxonSoft/AxxonOne/gpucache

5. Create the GPU_CACHE_DIR system variable,wherethevaluewillspecify the pathto the cachelocationfolder (see Creating system variables for the Axxon One server in Linux OS, Creating system variables for the Axxon One server in the failover mode in Linux OS). For example:

   Code Block
   language bash
   export GPU_CACHE_DIR="/opt/AxxonSoft/AxxonOne/gpucache"

6. Go to the /opt/AxxonSoft/DetectorPack/ folder:

   Code Block
   language bash
   cd /opt/AxxonSoft/DetectorPack

7. Run the following command:

   Code Block
   language bash
   ./RRGpuCacheGenerator

8. Select the required cache sets by specifying the y value:

   1. lpr recognition—recognition of a vehicle license plate;
   2. vehicle recognition—recognition of a vehicle.
      When you select the lpr recognition cache set, you need to specify the required countries in the numeric or alphabetic form. After you select all required countries, enter the done value.
      

      Expand
      title Expand the list of countries
      AM (Armenia) AR (Argentina) AT (Austria) AZ (Azerbaijan) BE (Belgium) BG (Bulgaria) BN (Brunei) BR (Brazil) BY (Belarus) CI (Ivory Coast) CN (China) CO (Colombia) CY (Cyprus) CZ (Czech Republic) DE (Germany) DK (Denmark) EE (Estonia) EG (Egypt) ES (Spain) FI (Finland) FR (France) GE (Georgia) GR (Greece) HR (Croatia) HU (Hungary) ID (Indonesia) IE (Ireland) IT (Italy) KG (Kyrgyzstan) KZ (Kazakhstan) LT (Lithuania) LU (Luxembourg) LV (Latvia) MD (Moldova) MM (Myanmar) MT (Malta) MX (Mexico) MY (Malaysia) NL (Netherlands) PE (Peru) PL (Poland) PT (Portugal) PA (Panama) PY (Paraguay) RO (Romania) RS (Serbia) RU (Russia) SE (Sweden) SI (Slovenia) SK (Slovakia) TJ (Tajikistan) TM (Turkmenistan) UA (Ukraine) UZ (Uzbekistan) UY (Uruguay) VN (Venezuela) VN (Vietnam) US (USA) LK (Sri Lanka) CL (Chile) TN (Tunisia)

   3. When you select the vehicle recognition cache set without lpr recognition, you must choose the territory where vehicle recognition is going to be performed:
      
      1. CIS countries—1;
      2. Other countries—2.
         
9. Select one of the accuracy and performance values by specifying the corresponding number:
   
   1. Medium accuracy, high performance—1, provides medium recognition accuracy and high performance of the GPU (less load on the GPU);
   2. High accuracy, medium performance—2, provides high recognition accuracy and medium performance of the GPU (high load on the GPU).
      
10. Specify the ID of the required GPU (see Selecting Nvidia GPU when configuring detectors).
11. Press Enter.

...

Optimizing the operation of the the License plate recognition RR and Vehicle recognition RR on RR—Parking detector on GPU is complete.

...