The Application of Optimized Pattern Learning Cultural Algorithm in Innovative Design of Environmental Art

Abstract

There are still some problems in the creative design of environmental art for standard cultural algorithms, such as low convergence accuracy and poor application effect. This paper proposes an environmental art creative design model based on a pattern learning cultural algorithm. Firstly, the genetic algorithm is used to provide population for the cultural algorithm's population space. Then the pattern extraction is used to obtain the characteristic information carried by the excellent individual, and the individuals in the population are organized regularly to perform pattern learning on this characteristic information, thus fully utilizing the guiding role of the excellent model. The algorithm incorporates the genetic algorithm into the framework of cultural algorithms and forms two main spaces based on GA: the main group space and the belief space. The main group space regularly organizes the best mode learning provided by the worst individual to the belief space in the evolution process, thus making full use of the characteristic information contained in a good individual greatly improves the convergence speed. Experimental results show that the algorithm is an effective algorithm to improve the performance of genetic algorithms.