The entire algorithm includes a car plate detection algorithm (using image segmentation) and a car plate recognition algorithm (CTC loss). The accuracy of this algorithm is 98.92% based on the data given.
The models can be downloaded from pan.baidu.com:
link:https://pan.baidu.com/s/1e8AtCfJ01fiu-vgJLRZ_ZQ code:s05b
The car plate training data is acquired from
https://github.com/detectRecog/CCPD
Please cite the paper if you are willing to use the dataset
@inproceedings{xu2018towards,
title={Towards End-to-End License Plate Detection and Recognition: A Large Dataset and Baseline},
author={Xu, Zhenbo and Yang, Wei and Meng, Ajin and Lu, Nanxue and Huang, Huan},
booktitle={Proceedings of the European Conference on Computer Vision (ECCV)},
pages={255--271},
year={2018}
}
The downloaded data should be put under the "./CCPD2019" directory.