Style Recognition of Calligraphic Chinese Characters Based on Morphological Convolutional Neural Network.

As an indispensable part of the excellent traditional Chinese culture, calligraphic Chinese characters have gradually evolved into different style types in the development process, which has raised the threshold for users to learn and appreciate calligraphy. With the development of deep learning technology, deep feature extraction technology based on convolutional neural network has made important breakthroughs in the task of calligraphy Chinese character style recognition. However, there are still problems such as lack of suitable datasets and easy loss of detailed feature information when extracting features, which lead to low accuracy of style recognition of calligraphy Chinese character. Therefore, this paper proposes a dilation pool subset based on morphological operators, and combines with residual block structure to build morphological convolutional neural network (MCNN) for calligraphy style recognition. The experimental results on 5 kinds of calligraphy Chinese character style datasets show that the recognition accuracy of the proposed method is 99.17%, and the recognition accuracy of cursive and running style is significantly improved by 4%–6% compared with other methods, which verifies the effectiveness of the proposed method for the style recognition of calligraphy Chinese characters. This study provides an effective solution for recognizing the style of Chinese characters in real scenes, and also has important research significance for broadening the application range of mathematical morphology.