https://github.com/sirius-ai/LPRNet_Pytorch

Pytorch Implementation For LPRNet, A High Performance And Lightweight License Plate Recognition Framework.
https://github.com/sirius-ai/LPRNet_Pytorch

ctc-loss license-plate-recognition lprnet plate-detection plate-recognition pytorch

Last synced: 5 days ago
JSON representation

Pytorch Implementation For LPRNet, A High Performance And Lightweight License Plate Recognition Framework.

• Host: GitHub
• URL: https://github.com/sirius-ai/LPRNet_Pytorch
• Owner: sirius-ai
• License: apache-2.0
• Created: 2019-05-04T15:05:14.000Z (about 5 years ago)
• Default Branch: master
• Last Pushed: 2023-06-21T13:37:25.000Z (about 1 year ago)
• Last Synced: 2024-03-12T22:31:12.603Z (4 months ago)
• Topics: ctc-loss, license-plate-recognition, lprnet, plate-detection, plate-recognition, pytorch
• Language: Python
• Homepage:
• Size: 19.6 MB
• Stars: 829
• Watchers: 10
• Forks: 216
• Open Issues: 68
• Metadata Files:
  • Readme: README.md
  • License: LICENSE

Lists

• awesome-stars - sirius-ai/LPRNet_Pytorch - Pytorch Implementation For LPRNet, A High Performance And Lightweight License Plate Recognition Framework. (Python)

README

        
# 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

* [pretrained_model](https://github.com/sirius-ai/LPRNet_Pytorch/tree/master/weights/)

# training and testing

1. prepare your datasets, image size must be 94x24.

2. base on your datsets path modify the scripts its hyperparameters --train_img_dirs or --test_img_dirs.

3. adjust other hyperparameters if need.

4. run 'python train_LPRNet.py' or 'python test_LPRNet.py'.

5. 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

1. [LPRNet: License Plate Recognition via Deep Neural Networks](https://arxiv.org/abs/1806.10447v1)

2. [PyTorch中文文档](https://pytorch-cn.readthedocs.io/zh/latest/)

# postscript

If you found this useful, please give me a star, thanks!