车牌评估在线查询“车牌查询”

Mobile-LPR 是一个面向移动端的准商业级车牌识别库，以NCNN作为推理后端，使用DNN作为算法核心，支持多种车牌检测算法，支持车牌识别和车牌颜色识别。

特点

超轻量，核心库只依赖NCNN，并且对模型量化进行支持多检测，支持SSD,MTCNN,LFFD等目标检测算法精度高，LFFD目标检测在CCPD检测AP达到98.9，车牌识别达到99.95%， 综合识别率超过99%易使用，只需要10行代码即可完成车牌识别易扩展，可快速扩展各类检测算法

算法流程

构建及安装

下载源码gitclonehttps://github.com/xiangweizeng/mobile-lpr.git准备环境安装opencv4.0及以上， freetype库安装cmake3.0以上版本，支持c++11的c++编译器，如gcc-6.3编译安装mkdirbuildcdbuildcmake..makeinstall

使用及样例

1.使用MTCNN检测

代码样例void test_mtcnn_plate(){
pr::fix_mtcnn_detector("../../models/float", pr::mtcnn_float_detector);
pr::PlateDetector detector = pr::IPlateDetector::create_plate_detector(pr::mtcnn_float_detector);

pr::fix_lpr_recognizer("../../models/float", pr::float_lpr_recognizer);
pr::LPRRecognizer lpr = pr::float_lpr_recognizer.create_recognizer();
Mat img = imread("../../image/plate.png");

ncnn::Mat sample = ncnn::Mat::from_pixels(img.data, ncnn::Mat::PIXEL_BGR, img.cols, img.rows);
std::vector objects;
detector->plate_detect(sample, objects);
lpr->decode_plate_infos(objects);for(auto pi : objects)
{
cout <<"plate_no: "<< pi.plate_color << pi.plate_no <<" box:"<< pi.bbox.xmin <<","<< pi.bbox.ymin <<","<< pi.bbox.xmax <<","<< pi.bbox.ymax <<","<< pi.bbox.score << endl;
}
}效果示例:

2.使用LFFD检测

代码样例void test_lffd_plate()
{
pr::fix_lffd_detector("../../models/float", pr::lffd_float_detector);
pr::PlateDetector detector = pr::IPlateDetector::create_plate_detector(pr::lffd_float_detector);

pr::fix_lpr_recognizer("../../models/float", pr::float_lpr_recognizer);
pr::LPRRecognizer lpr = pr::float_lpr_recognizer.create_recognizer();
Mat img = imread("../../image/plate.png");

ncnn::Mat sample = ncnn::Mat::from_pixels(img.data, ncnn::Mat::PIXEL_BGR, img.cols, img.rows);
std::vector objects;
detector->plate_detect(sample, objects);
lpr->decode_plate_infos(objects);for(auto pi : objects)
{
cout <<"plate_no: "<< pi.plate_color << pi.plate_no <<" box:"<< pi.bbox.xmin <<","<< pi.bbox.ymin <<","<< pi.bbox.xmax <<","<< pi.bbox.ymax <<","<< pi.bbox.score << endl;
}
}效果示例:

3.使用SSD检测

代码样例void test_ssd_plate()
{
pr::fix_ssd_detector("../../models/float", pr::ssd_float_detector);
pr::PlateDetector detector = pr::IPlateDetector::create_plate_detector(pr::ssd_float_detector);

pr::fix_lpr_recognizer("../../models/float", pr::float_lpr_recognizer);
pr::LPRRecognizer lpr = pr::float_lpr_recognizer.create_recognizer();
Mat img = imread("../../image/manys.jpeg");

ncnn::Mat sample = ncnn::Mat::from_pixels(img.data, ncnn::Mat::PIXEL_BGR, img.cols, img.rows);
std::vector objects;
detector->plate_detect(sample, objects);
lpr->decode_plate_infos(objects);for(auto pi : objects)
{
cout <<"plate_no: "<< pi.plate_color << pi.plate_no <<" box:"<< pi.bbox.xmin <<","<< pi.bbox.ymin <<","<< pi.bbox.xmax <<","<< pi.bbox.ymax <<","<< pi.bbox.score << endl;
}
}效果示例:

4.使用量化模型

代码样例void test_quantize_mtcnn_plate(){
pr::fix_mtcnn_detector("../../models/quantize", pr::mtcnn_int8_detector);
pr::PlateDetector detector = pr::IPlateDetector::create_plate_detector(pr::mtcnn_int8_detector);

pr::fix_lpr_recognizer("../../models/quantize", pr::int8_lpr_recognizer);
pr::LPRRecognizer lpr = pr::int8_lpr_recognizer.create_recognizer();
Mat img = imread("../../image/plate.png");

ncnn::Mat sample = ncnn::Mat::from_pixels(img.data, ncnn::Mat::PIXEL_BGR, img.cols, img.rows);
std::vector objects;
detector->plate_detect(sample, objects);
lpr->decode_plate_infos(objects);for(auto pi : objects)
{
cout <<"plate_no: "<< pi.plate_color << pi.plate_no <<" box:"<< pi.bbox.xmin <<","<< pi.bbox.ymin <<","<< pi.bbox.xmax <<","<< pi.bbox.ymax <<","<< pi.bbox.score << endl;
}
}效果示例:

后续工作

添加更优的算法支持优化模型，支持更多的车牌类型，目前支持普通车牌识别，欢迎各位大神提供更好的模型优化模型，更高的精度添加Android 使用实例性能评估