Automatic generation of traditional patterns and aesthetic quality evaluation technology

Abstract

As an important element for the continuation and development of modern design, traditional Chinese decorative patterns are to satisfy people's emotional dependence on nationality and tradition. At present, traditional patterns are mostly completed by professional designers. Due to the high dependence of the entire design process on labor, the design e ciency is di cult to quickly improve to meet the explosive growth of business needs. However, the large amount of material data on the Internet and the development of arti cial intelligence technology provide an opportunity to solve this development bottleneck. This paper mainly studies the design and implementation of the automatic generation of traditional pattern image layout based on the generation of confrontation network and the use of aesthetic evaluation to score the quality of the generated traditional pattern image. And the scoring results are fed back to the traditional pattern layout generation network, so as to improve the quality of the generated traditional pattern layout pictures, so that the entire network structure can adapt to different scene requirements. The results show that the SSIM value of the structure similarity obtained by the image of the traditional pattern automatically generated by the model in this paper is larger and closer to 1 than the value of the SSIM obtained by other methods, indicating that the image of the traditional pattern automatically generated by the model has the highest aesthetic quality.

Research on the Design of Zhuang Brocade Patterns Based on Automatic Pattern Generation

“…AGPG models generate multi-purpose camouflage patterns [11], IFS and curve functions simulate floral patterns [3], and color transfer methods create colored mandalas [12]. Adversarial networks and aesthetics guide auto-gen pattern design [13].…”