Application of Deep Convolution Neural Network Algorithm in Detecting Traditional Calligraphy Characters

Calligraphy is an important part of Chinese culture, and calligraphy detection is of great significance. At present, there are still some challenges in the detection of ancient calligraphy. In this paper, we are interested in the calligraphy detection problem with a focus on the calligraphy character boundary. We choose High-Resolution Net (HRNet) as the calligraphy character feature extraction backbone network to learn reliable high-resolution representations. Then we use the scale prediction branch and the spatial information prediction branch to detect the calligraphy character region and categorize the calligraphy character and its boundaries. In this process we use the channel attention mechanism and the feature fusion method to improve detection effectiveness. Finally, we compare our result with the result that is detected without boundary. The comparison proves the superiority of our method, and our method can accurately detect each calligraphy character.