Handwritten Character Recognition Based on 13-point Feature of Skeleton and Self-Organizing Competition Network

This article is about the recognition for handwritten characters. Firstly, we use some conventional methods to preprocess the images and introduce the 13-point feature of skeleton method to extract the data that containing feature of the handwritten character. Secondly, we use the data to train the self-organizing competition network. In the end, we test the net and conclude that this method has a good performance at handwritten character recognition.