Fuzzy-Based Lion Pride optimization for Grayscale Image Segmentation

Abstract

In this paper, we propose a novel image segmentation algorithm, Fuzzy-Based Differential Lion Pride optimization (FDLPO). We combine the concept of the Fuzzy C Means (FCM) and an improved Lion Pride optimization (DLPO) with a new crossover strategy which is inspired by Differential Evolution (DE). First, we search for optimum clusters by fuzzy membership function and LPO which has better performance than most of the other algorithms. Then, the crossover mechanism of the differential evolution is used to improve the diversity of offspring and speed up convergence in lion pride optimization. FDLPO becomes more efficient as it overcomes the drawbacks of FCM and LPO which does not depend on the choice of initial cluster centers and it performs better in terms of convergence, time complexity and robustness. The experiments with DLPO, LPO and PSO are executed over some benchmark test functions. Meanwhile, the efficiency of FDLPO is proven by both quantitative and qualitative measures.