Research Article
Electric Energy Meter Information Recognition System Based on Deep Learning
@INPROCEEDINGS{10.1007/978-3-031-23902-1_26, author={Shuai Gao and Bo Ning and Nan Jiang}, title={Electric Energy Meter Information Recognition System Based on Deep Learning}, proceedings={Mobile Multimedia Communications. 15th EAI International Conference, MobiMedia 2022, Virtual Event, July 22-24, 2022, Proceedings}, proceedings_a={MOBIMEDIA}, year={2023}, month={2}, keywords={OCR technology CTPN network CRNN network Electric energy meter information recognition}, doi={10.1007/978-3-031-23902-1_26} }- Shuai Gao
Bo Ning
Nan Jiang
Year: 2023
Electric Energy Meter Information Recognition System Based on Deep Learning
MOBIMEDIA
Springer
DOI: 10.1007/978-3-031-23902-1_26
Abstract
A new type of electric energy meter information recognition system based on deep learning is proposed. The system is mainly divided into OCR character recognition system and electric meter information verification system. OCR character recognition system mainly includes two parts: character detection and character recognition. The text detection uses the CTPN model, and the text recognition uses the CRNN network in deep learning for recognition, and then uses the CTC loss function for sequence processing to improve the accuracy of text recognition. Through the RCTW-17 data set training, an OCR text recognition system with high accuracy, strong stability and fast speed is obtained. The identified results are automatically checked with the information in the background database, and finally an electric energy meter information identification system is obtained. The verification of the RCTW-17 data set and the actual photo identification of the electric energy meter prove the effectiveness of this method.
