LPRNet_Pytorch
Pytorch Implementation For LPRNet, A High Performance And Lightweight License Plate Recognition Framework.
完全适用于中国车牌识别（Chinese License Plate Recognition）及国外车牌识别！
目前仅支持同时识别蓝牌和绿牌即新能源车牌等中国车牌，但可通过扩展训练数据或微调支持其他类型车牌及提高识别准确率！
dependencies
- pytorch >= 1.0.0
- opencv-python 3.x
- python 3.x
- imutils
- Pillow
- numpy
pretrained model
training and testing
- prepare your datasets, image size must be 94x24.
- base on your datsets path modify the scripts its hyperparameters --train_img_dirs or --test_img_dirs.
- adjust other hyperparameters if need.
- run 'python train_LPRNet.py' or 'python test_LPRNet.py'.
- if want to show testing result, add '--show true' or '--show 1' to run command.
performance
- personal test datasets.
- include blue/green license plate.
- images are very widely.
- total test images number is 27320.
| size | personal test imgs(%) | inference@gtx 1060(ms) |
|---|---|---|
| 1.7M | 96.0+ | 0.5- |
References
postscript
If you found this useful, please give me a star, thanks!