Intelligent Image Analysis and Recognition Method for Art Design Majors

Abstract

Art design major is a relatively important course in college teaching. It involves a wide range of directions. Advertising art, landscape art, interior design, etc. are closely related to people’s lives. Art design has appeared in all aspects of people’s lives. However, a new art design program is time-consuming and human resources for art design. Different art designs will contain relatively similar characteristics, which can alleviate many difficulties for art designers. However, it is also a relatively difficult task to discover the relationship between the characteristics of the art and design only by artificial means. Image recognition technology can assist designers to discover and find the relationship between artworks, and these related features can assist designers to design. In this study, an intelligent image recognition method for intelligent art design is designed using the VB-CNN-GRU method. It can identify patterns, shapes, color matching, and text features of artistic design products. The research results show that the VB-CNN-GRU method can accurately complete the intelligent image recognition task of art design major. The VB-CNN-GRU method has specifically higher accuracy in art design image recognition than the single VB-CNN method. The maximum prediction error of VB-CNN-GRU in art design image recognition is only 2.37%. For the four characteristics of art design, it can better assist designers to complete related designs.