Neural Networks that Learn to Discriminate Similar Kanji Characters
Yoshihiro Mori, Kazuhiko Yokosawa · Neural Information Processing Systems · 1988
A neural network is applied to the problem of recognizing Kanji characters. Using a back propagation network learning algorithm, a three-layered, feed-forward network is trained to recognize similar handwritten Kanji characters. In addition, two new methods are utilized to make training effective. The recognition accuracy was higher than that of conventional methods. An analysis of connection weights showed that trained networks can discern the hierarchical structure of Kanji characters. This strategy of trained networks makes high recognition accuracy possible. Our results suggest that neural networks are very effective for Kanji character recognition.