A Novel Chaotic Artificial Gorilla Troops Optimizer and Its Application for Fundus Images Segmentation

This paper introduces an improved version of Gorilla Troops Optimizer (GTO) based on Chaotic maps, namely chaotic Gorilla Troops Optimizer (CGTO). GTO like other swarm intelligence algorithms suffers from stagnation in the local optima problem during the optimization process and premature convergence. The proposed CGTO is used to tackle these problems and thus boost the performance of the standard GTO. The performance of the proposed CGTO is tested and evaluated to find optimal solutions for global optimization and multilevel thresholding optimization problems. Three chaotic maps are adopted and evaluated. These maps are Circle, Gauss, and Tent chaotic maps. The experimental results revealed that the proposed CGTO is superior compared with other swarm optimization algorithms. Moreover, the results are validated quantitatively and qualitatively for fundus images. The simulation results showed that the proposed can find dominant regions compared with the original GTO.