All Projects → Fanghc95 → TensorRT-LPR

Fanghc95 / TensorRT-LPR

Licence: other
车牌识别,基于HyperLPR实现,修改模型调用方法,使用caffe+tensorRT实现GPU加速,修改了车牌检测模型

Programming Languages

C++
36643 projects - #6 most used programming language
Cuda
1817 projects
CMake
9771 projects

Projects that are alternatives of or similar to TensorRT-LPR

Mobilenet Yolo
A caffe implementation of MobileNet-YOLO detection network
Stars: ✭ 825 (+5792.86%)
Mutual labels:  caffe, yolo
Caffe Yolov3 Windows
A windows caffe implementation of YOLO detection network
Stars: ✭ 210 (+1400%)
Mutual labels:  caffe, yolo
Pytorch Caffe Darknet Convert
convert between pytorch, caffe prototxt/weights and darknet cfg/weights
Stars: ✭ 867 (+6092.86%)
Mutual labels:  caffe, yolo
Jetson Inference
Hello AI World guide to deploying deep-learning inference networks and deep vision primitives with TensorRT and NVIDIA Jetson.
Stars: ✭ 5,191 (+36978.57%)
Mutual labels:  caffe, tensorrt
pnn
pnn is Darknet compatible neural nets inference engine implemented in Rust.
Stars: ✭ 17 (+21.43%)
Mutual labels:  yolo, tensorrt
Maskyolo caffe
YOLO V2 & V3 , YOLO Combined with RCNN and MaskRCNN
Stars: ✭ 101 (+621.43%)
Mutual labels:  caffe, yolo
Deepdetect
Deep Learning API and Server in C++14 support for Caffe, Caffe2, PyTorch,TensorRT, Dlib, NCNN, Tensorflow, XGBoost and TSNE
Stars: ✭ 2,306 (+16371.43%)
Mutual labels:  caffe, tensorrt
Caffe2 Ios
Caffe2 on iOS Real-time Demo. Test with Your Own Model and Photos.
Stars: ✭ 221 (+1478.57%)
Mutual labels:  caffe, yolo
ros-yolo-sort
YOLO v3, v4, v5, v6, v7 + SORT tracking + ROS platform. Supporting: YOLO with Darknet, OpenCV(DNN), OpenVINO, TensorRT(tkDNN). SORT supports python(original) and C++. (Not Deep SORT)
Stars: ✭ 162 (+1057.14%)
Mutual labels:  yolo, tensorrt
tensorrt-ssd-easy
No description or website provided.
Stars: ✭ 32 (+128.57%)
Mutual labels:  caffe, tensorrt
darknet2caffe
Conversion of yolo from DarkNet to Caffe
Stars: ✭ 25 (+78.57%)
Mutual labels:  caffe, yolo
Scaled-YOLOv4-TensorRT
Got 100fps on TX2. Got 500fps on GeForce GTX 1660 Ti. If the project is useful to you, please Star it.
Stars: ✭ 169 (+1107.14%)
Mutual labels:  yolo, tensorrt
YOLOX
YOLOX is a high-performance anchor-free YOLO, exceeding yolov3~v5 with MegEngine, ONNX, TensorRT, ncnn, and OpenVINO supported. Documentation: https://yolox.readthedocs.io/
Stars: ✭ 6,570 (+46828.57%)
Mutual labels:  yolo, tensorrt
R2CNN
caffe re-implementation of R2CNN: Rotational Region CNN for Orientation Robust Scene Text Detection
Stars: ✭ 80 (+471.43%)
Mutual labels:  caffe
caffe-mobilenet-v3
Caffe Implementation of MobileNets V3
Stars: ✭ 127 (+807.14%)
Mutual labels:  caffe
person-detection
TensorRT person tracking RFBNet300
Stars: ✭ 30 (+114.29%)
Mutual labels:  tensorrt
aikanshe
使用SpingBoot、OpenCV、yolo开发舌诊小程序
Stars: ✭ 22 (+57.14%)
Mutual labels:  yolo
yolo-deepsort-flask
Target detection and multi target tracking platform based on Yolo DeepSort and Flask.
Stars: ✭ 29 (+107.14%)
Mutual labels:  yolo
caffe
This fork of BVLC/Caffe is dedicated to supporting Cambricon deep learning processor and improving performance of this deep learning framework when running on Machine Learning Unit(MLU).
Stars: ✭ 40 (+185.71%)
Mutual labels:  caffe
tensorflow-example-java
This is a Tensorflow Java example application what uses YOLOv2 model and Gradle for build and dependency management.
Stars: ✭ 49 (+250%)
Mutual labels:  yolo

TensorRT-LPR

车牌识别 该工程车牌检测技术参考YOLOv3,识别技术参考HyperLPR

基于Retinaface的车牌检测及关键点检测项目可参考Plate-Landmarks-detection

工程基于HyperLPR/prj-Linux修改:

1.修改车牌检测方法,改为yolov3实现;

2.源工程使用opencv-dnn模块对模型进行部署运行,改为caffe+tensorRT部署,可调用GPU进行加速

使用

工程环境:

  1. caffe
  2. OpenCV 3.4.2
  3. TensorRT 6.0.1.5

依赖三方工程:

1.tensorRTWrapper:用于部署YOLOv3等模型;

2.darknet:YOLOv3车牌检测模型训练;

3.darkner2caffe:darknet模型转为caffe模型。

模型:

  1. yolov3对象检测模型百度(提取码:vm66 )

  2. 车牌垂直边界回归模型

  3. e2e车牌识别模型

yolo对象检测模型可使用darknet自行训练,数据集参考CCPD

部署时需要使用darkner2caffe将darknet模型转换为caffe模型(注意工程中yolov3输入尺寸为416)

车牌边界回归以及识别模型来自HyperLPR/prj-Linux

开始:

git clone [email protected]:Fanghc95/TensorRT-LPR.git
#编辑CMakeList.txt配置CUDA,tensorRT,opencv等
mkdir build&&cd build
cmake ../
make -j8
#构建对象检测模型engine
./testPlate 0
#构建车牌识别模型engine
./testPlate 1
#部署并测试单张图片
./testPlate 2 [img_path]

改进方面:

  1. 垂直边界回归没能用到tensorRT,我在部署时没能跑通,大佬们可以继续改进

  2. 字符识别部分我用的是开源模型,效果较好但称不上100%完美,因为没接触过OCR没有进行改进(其实也莫得数据)

  3. 公开数据集CCPD虽数据量大但是场景单一,检测部分训练还需要额外数据进行优化

Note that the project description data, including the texts, logos, images, and/or trademarks, for each open source project belongs to its rightful owner. If you wish to add or remove any projects, please contact us at [email protected].