Testing and Improvement of OCR Recognition Technology in Export-Oriented Chinese Dictionary APP

Qingpei Yang, Xinrui Wan, Hongjun Chen, Mengjiao Wu, Yao Wu, Jiyun Qiao

Abstract: With the increasing number of Chinese learners each year, the international influence of the Chinese language has also grown. Consequently, Chinese auxiliary learning tools should align more closely with the needs of present-day learners. This paper focuses on a typical Chinese dictionary app and conducts tests and comparisons of the optical character recognition (OCR) function. Specifically, it compares the OCR function of a widely used Chinese learning dictionary with the more advanced and sophisticated mobile built-in OCR recognition technology available today. By identifying the technical gap, the paper proposes a design that incorporates deep learning techniques to enhance the built-in OCR function of the dictionary. This improvement aims to significantly enhance recognition accuracy in natural scenarios and ultimately enhance the user experience of the Chinese learning dictionary app.