Progressive calligraphy stroke extraction
Fang Chen, Xiafen Zhang · Proceedings of the 2022 2nd International Conference on Control and Intelligent Robotics · 2022
Calligraphy is a valuable traditional culture in China. The study of calligraphy can be used in many aspects. At present, there is little research on calligraphy stroke extraction, and there is a lack of practical and effective automatic calligraphy stroke extraction algorithm. In this paper, a stroke extraction algorithm for calligraphy image is proposed, which uses multiple rounds of iterative and progressive method to extract strokes. It is mainly divided into four steps: segment skeleton acquisition, stroke skeleton generation, stroke acquisition with width information and improved stroke skeleton acquisition. Firstly, the algorithm obtains the skeleton of each pen segment of calligraphy, judges the pen segment type, and combines the pen segment skeleton to generate the stroke skeleton. Determine the stroke intersection type, determine the inflection point combined with the contour characteristics and the center point of the intersection area, and complete the stroke contour with Bezier curve. Then the connected part of the strokes is processed to obtain the strokes with width information. Finally, the extracted strokes are refined to obtain the improved stroke skeleton. The experiments show that the proposed algorithm has achieved good results in the stroke extraction of calligraphy image, and the improved stroke skeleton can better reflect the real trend of strokes, which can lay a foundation for the research of calligraphy based on stroke level.