License Plate Recognition🔗

Module License Plate Recognition is intended to recognize state license plates of cars with saving information about the time and date of recognition, the license plate of the car, as well as links to the corresponding video frame in the archive.

Capabilities of the module🔗

Recognition by attributes🔗

Recognition of license plate with the following information:

  • Vehicle registration number

  • Country code

  • The color of the license plate

  • Vehicle brand

  • Vehicle color

  • Vehicle type

License plate recognition for the countries listed below:

Armenia (AM), Azerbaijan (AZ), Belarus (BY), China (CH), Georgia (GE), Great Britain (GB), India (IN), Indonesia (ID), Kazakhstan (KZ), Kyrgyzstan (KG), Moldova (MD), Russia (RU), Tajikistan (TJ), Turkey (TR), Uzbekistan (UZ).

Warning

Correct country code recognition is only possible for countries from the supported list.

Note

The module can correctly recognize license plates from countries not included in the supported list, provided that they contain only Latin alphabet characters and Arabic numerals. However, the accuracy of recognition of such license plates is not guaranteed.

The module operates at vehicle speeds up to 270 km/h.

The following vehicle brands are defined:

Audi, BAIC, BMW, Buick, CNHTC, Changan (Chana), Chery (Chirey), Chevrolet, Citroen, Daewoo, Daihatsu, Dodge, FAW, FIAT, Ford, Freightliner, GAZ, Geely, GreatWall, Hino, Honda, Hyundai, Infiniti, Isuzu, Iveco, JAC, Jaguar, Jeep, KIA, KamAZ, KrAZ, Lada (VAZ), Landrover, Lexus, Lifan, MAN, MAZ, MAZDA, Mercedes-Benz, MINI, Mitsubishi, Nissan, Opel, Paccar, Peugeot, Porsche, Renault, SAIC, Scania, Shacman, Skoda, Subaru, Suzuki, Tatra, Tesla, Toyota, UAZ, Ural (UralAZ), Vauxhall, Volkswagen, Volvo, ZIL.

The vehicle color is determined from a preset list:

  • White

  • Yellow

  • Green

  • Red

  • Brown

  • Silver

  • Blue

  • Gray

  • Black.

Note

When you select to display the second color in the settings, both colors are shown with their priority.

The vehicle is classified into the following types:

  • Passenger car

  • Truck

  • Light truck

  • Motorcycle

  • Bus

  • Construction equipment.

Note

  • Vans are of the Light truck type.

  • Minibuses are of the Bus type.

When recognizing a license plate the Recognized license plates event is generated.

Accuracy of recognizing vehicle attributes:

  • Brand - 94%.

  • Color detection:

    • One color - 90%

    • Two colors - 94%

Vehicle type - 92%.

Warning

The specified accuracy metrics were obtained using a limited dataset representative of the module's operational conditions.

Warning

Enabling the Identify the type, color and model of the vehicle option increases the system load by 5-10%.

Counting vehicles in a parking lot🔗

The following parking tracking features are available:

  • Counting the number of vehicles in the parking lot

  • Tracking the duration of a vehicle's presence in the parking lot

Important

To track parking lots, enable and configure the corresponding option in the Eocortex Configurator application. For more information, see Parking spots.

When tracking parking lots, the following events are generated: Parking vehicle counter has been changed and Parking lot occupancy has changed.

For more information on using this option, see Recognized license plates panel.

Reports🔗

The following reports are available in the module and similar Automatic reports:

Additional information🔗

To automate the extraction of recognized license plate lists, you can use the capabilities of Eocortex API.

Interaction with barriers can be implemented through signal outputs of IP cameras or other network devices supported by Eocortex.

Warning

The country code is not displayed in reports and the camera cell. It can only be obtained through Eocortex API. At the same time, the country code can be used in automation scenarios under the Recognized license plates event conditions.

Note

When switching to this module from other license plate recognition modules, the previously recorded events will be preserved.

Note

The license plate database is the common for all license plate recognition modules.

The module can use NVIDIA video cards to reduce CPU load.

Compatibility with other modules.🔗

Modules that can be enabled simultaneously with the License Plate Recognition module on a single camera:

  • Loud sound detection

  • Fire and Smoke Detection

  • Abandoned object detection module

  • Face Mask Detection

  • Sabotage Detection

  • Crowd Monitoring

  • Personnel Activity Monitoring

  • Shelf Fullness Check

  • Uniform Detection

  • Face detection

  • Emergency Vehicle Detection

  • Counting People in Queue

  • Object Classification and Counting1

  • People Counting1

  • Unique Visitor Counting

  • Search for Objects

  • Frame Area Blurring

  • Face Recognition (Complete)1

  • Face Recognition (Light)1

  • Traffic density heat map

  • Tracking

1 With simultaneous use, a module selection is available for displaying the analysis results in the Eocortex Client application.

Not recommended for use with modules:

  • Auto Zoom

Incompatible with modules:

  • Fisheye dewarping module

  • License Plate Recognition (Complete)

Use of the module in the client applications🔗

Related references