Recognition Model Based On Residual Networks For Cursive Hanja Recognition

With the development of algorithms and computer skills, deep learning using CNN (convolutional neural network) has been applied to various fields, especially in image processing field. In this paper, we designed an improved model based on ResNet with CNN structure, and learned the database. The Chaucer Database used in the experiment consisted of 824 Chinese characters among the Chinese characters registered in the Dictionary of Old Document Type Usage Dictionary of the Korean Studies Data Portal and total of 240,000 data. The experiment used 10-fold cross-validation. The ResNet-based percussive network used in the experiments showed an average of 94.7% top-1 accuracy in the post hoc classification test.