Multi-threshold Image Segmentation by Improved Lion Swarm Optimization Algorithm

Abstract

Abstract In this paper we propose an improved Lion Swarm Optimization (ILSO) algorithm for multi-threshold image segmentation. We introduce the global search method of Artificial Bee Colony optimization (ABC) algorithm and revise the updating function of LSO algorithm to improve the global and local search performance of LSO algorithm. We introduce a sign to record the number of times an individual falls into the local optimum, and attenuate it so that the algorithm can jump out of the local optimum more quickly in the early stage and accelerate the convergence speed in the later stage. The maximum inter-class variance criterion is selected as the fitness function to solve the Multi-threshold image segmentation problem by ILSO. Experiment results show this algorithm can obtain ideal image segmentation result. And when the dimension of the problem is higher, the advantage of the improved Lion Swarm Optimization algorithm proposed in this paper is more obvious.