Dynamic Reconstruction Algorithm of Calligraphy Characters Based on Self-organizing Mapping

Abstract

Due to the growth of society and the steady progress of economic construction, people are no longer only satisfied with the material needs of life, but more begin to pursue spiritual satisfaction. As a traditional art in China, China calligraphy has renewed its vitality. The rapid growth of artificial intelligence (AI) provides a new storage medium for calligraphy works and brings new ideas for the inheritance and dissemination of calligraphy art. The depth image contains the depth information of the scene. It is precisely because the depth information contained in the depth image represents the surface geometry of the objects in the scene that the depth image can directly use the 3D information of the geometry of these objects to reconstruct the scene in three dimensions. In this article, a dynamic emphasis algorithm of calligraphy characters based on self-organizing mapping (SOM) and CAD is proposed, and the calligraphy skeleton is extracted by SOM algorithm to ensure the continuity and smoothness of the calligraphy skeleton, so as to better reproduce the calligraphy characters in three dimensions. The comprehensive experimental results show that SOM model has a good application effect on the dynamic reproduction of calligraphy characters, and the image feature recognition accuracy is high, and the optimized image effect is obviously better than the comparison algorithm.