Extraction and Reuse of Pattern Configuration for Personalized Customization of Cantonese Porcelain Based on Artificial Intelligence

摘要

To solve the problems of inefficient learning caused by the complexity, fragmentation and lack of personalization of traditional handicraft learning resources. Based on the method of pattern configuration extraction and reuse, this paper takes the innovative design of Canton Porcelain pattern as an example. For the first time, sorting the knowledge of Cantonese Porcelain according to different cultural attributes (labels), to establish the knowledge base and pattern sample base of Canton Porcelain, and build a semantic relationship between the them, recommend Cantonese Porcelain elements that meet users' needs through semantic search. Then, shape context was used to extract patterns of Cantonese Porcelain, and topological methods was combined to establish the configuration rules of patterns. Shape grammar based on character encoding were improved in the process of generation of personalized customization patterns of Cantonese Porcelain, used to describe the transformation of shapes during pattern filling, thereby generation of personalized customization patterns of Cantonese Porcelain based on element extraction will be completed. Develop a personalized customization system for Cantonese Porcelain crafts, and tests the method feasible through an application example.