A Lightweight Handwriting Recognition System Based on an Improved Convolutional Neural Network
Yingmiao Jia, Yanfei Zhao, Yan Zhang, Shurui Fan · 2020
For decades, how to recognize handwritten characters quickly and efficiently has been a difficult problem in research. Handwritten character recognition has been widely used in our life. The existing handwriting recognition system relies on the computer and can't recognize the text in real time. In this paper, a new improved handwritten character recognition method of convolutional neural network is proposed, which adds Batch Normalization Layer and Residual Network Structure to the general convolutional neural network, and adopts multi-scale prediction method. The model is proved to have higher recognition rate by test set. On the computer side, an improved convolutional neural network was built through Keras to train the EMNIST data set and obtain the model. After migrating the model to the mobile terminal, an off-line real-time handwritten character recognition system was developed.