An improved whale optimization algorithm for dinosaur lantern festival color image multilevel thresholding segmentation

The whale optimization algorithm has the shortcomings of precocity and insufficient diversity when it is used to segment the color lantern image. A chaotic levy whale optimization algorithm is proposed to search the optimal Kapur s entropy and Renyi entropy. In this algorithm, particles are used to represent a feasible one-dimensional multi threshold vector, and chaos initialization strategy is adopted to improve the randomness of the algorithm. Based on the whale optimization algorithm, Levy flight strategy is used to enhance the diversity and global search ability of the algorithm. The simulation results show that the proposed algorithm is better than the other five algorithms. In addition, the algorithm using Renyi entropy has better performance than Kapur s entropy in terms of Signal to Noise Ratio (PSNR), Feature Similarity index (FSIM), and best fitness. It also provides a reference for the application of swarm intelligence algorithm in intangible cultural heritage image segmentation.