Detecting Chinese calligraphy style consistency by deep learning and one-class SVM

When beginners practice Chinese calligraphy, they often copy from ancient calligraphic works and try to imitate the style as closely as possible. However there are inevitably some characters whose styles are not correctly followed. Thus we are motivated to detect the style consistency of all written characters in one practice. With the styles extracted by using stacked autoencoders of deep neural network model, we discriminate correctly styled and alien styled characters using a trained one-class support vector machine. Thus we can pick out those outliers. The proposed algorithm reaches satisfactory results. The algorithm can also be applied to other image style detection problems.