Deep Convolutional Network for Handwritten Chinese Character Recognition

Code

• github.com/itanghiu/OnlineHanziRecognizer_TF2_5
  tf★ 0

Abstract

In this project we explored the performance of deep convolutional neural network on recognizing handwritten Chinese characters. We ran experiments on a 200-class and a 3755-class dataset using convolutional networks with different depth and filter numbers. Experimental results show that deeper network with larger filter numbers give better test accuracy. We also provide a visualization of the learned network on the handwritten Chinese characters.

Reproductions