Bank card number recognition system based on deep learning
Gege Sun, Fucheng You · Proceedings of the 2020 4th International Conference on Electronic Information Technology and Computer Engineering · 2020
With the widespread application of mobile payment, identifying the bank card number can help us bind the bank card quickly. The angle of bank card shooting, the light and shade of light, and the background colour of cards from different companies will affect card number recognition accuracy. Therefore, this paper designs a bank card number recognition system, which mainly includes three parts: card number location, character recognition and GUI interface design. Firstly, the improved EAST algorithm is proposed to locate the card number, and then the neural network training and character recognition are carried out. In this process, the CRNN-BLSTM-CTC architecture of the mainstream OCR algorithm is mainly used. Finally, the GUI is used to demonstrate the results. The experimental results show that the system significantly improves the accuracy and efficiency of bank card number recognition.