Calligraphy Font Recognition Algorithm based on Improved DenseNet network

The role of calligraphy fonts in Chinese character calligraphy is very important, but there is a problem that the font cannot be determined. The traditional DenseNet network cannot solve the problem of inaccurate calligraphy font recognition in Chinese character calligraphy, moreover. Therefore, this paper proposes an improved DenseNet network for accurate calligraphy font analysis. Firstly, the computer is used to identify and analyze the font, and the index is divided according to the requirements of the calligraphy font to reduce the calligraphy font in the interfering factor. Then, the computer recognizes the results of the Chinese character calligraphy font, forms the calligraphy font scheme, and conducts the calligraphy font result Comprehensive analysis. MATLAB simulation shows that the calligraphy font of DenseNet network for Chinese character calligraphy is improved under certain evaluation criteria The recognition accuracy and calligraphy font recognition time are better than those of the traditional DenseNet network.