Multi-level threshold segmentation based on LSHADE
Guang Yang, ZhaoGuang Liu, Zong-Na Zhu · 2021
In order to solve the problems in traditional multi-threshold image segmentation methods, a Multi-threshold image segmentation method based on LSHADE is proposed. On the basis of SHADE, LSHADE uses a linear population size reduction method to continuously reduce the population size to improve the performance of the algorithm. First, perform multi-threshold image segmentation based on the between-class variance and entropy, and compare the segmentation results of the two methods. Then using PSNR and objective function values as evaluation criteria, comparing LSHADE with other algorithms. The results show that the segmentation speed and segmentation accuracy have been improved.