ANPR is an intelligent license plate recognition based on deep learning, which supports 14 types of Chinese license plates.

• The most complete license plate recognition algorithm supports 14 Chinese license plate types.

• The project is developed in pure Java to avoid service instability caused by using Python.

• Everyone is welcome to contribute code. If you think the project is not bad, please give it a star.

Project Description

1. spring boot

2,onnx

1,Chinese_license_plate_detection_recognition

project documentation

JAVA client

• Java dependency, not published to the central warehouse, need to compile and publish to the private warehouse

<dependency>
    <groupId>com.visual.open.anpr</groupId>
    <artifactId>open-anpr-client</artifactId>
    <version>1.0.0</version>
</dependency>

Use the restful interface:docs-1.0.0

project deployment

1、直接docker部署项目：
docker run -d --name "open-anpr" -p 56790:8080 divenswu/open-anpr:1.0.0

2、若想禁用swagger：
docker run -d --name "open-anpr" -p 56790:8080 -e VISUAL_SWAGGER_ENABLE=false divenswu/open-anpr:1.0.0

3、若想挂载日志目录到本地：
docker run -d --name "open-anpr" -p 56790:8080 -v open-anpr/logs:/app/open-anpr/logs divenswu/open-anpr:1.0.0

4、服务访问：
  swagger接口文档： http://127.0.0.1:56790/doc.html

• The project is compiled and packaged as a docker image

1、java版本最低为：1.8；安装maven编译工具。安装docker。
2、克隆项目
  git clone https://gitee.com/open-visual/open-anpr.git
3、项目打包
   cd open-anpr && sh scripts/docker_build.sh

project demo

License plate recognition (better mobile phone experience)