Research on Illustration Art Design Models Based on Operator-Optimized Genetic Algorithms

This paper addresses the challenges of low search efficiency and high complexity associated with standard genetic algorithms in illustration art design. We introduce a new model that integrates operator optimization and clustering-enhanced genetic algorithms. First, we dynamically modify the crossover and mutation probabilities during the genetic algorithm's execution based on population characteristics, which improves search efficiency.Additionally, we incorporate the K-medoids algorithm to refine clustering within the genetic framework, utilizing a cost function to assess clustering quality and thereby reduce the algorithm's complexity. Finally, we apply our enhanced genetic algorithm to the design of random illustrations. Simulation results demonstrate that our proposed algorithm achieves higher levels of artistic quality and innovation in illustration art design.