Improvement Cat Swarm Optimization for Efficient Motion Estimation

Abstract

Cat swarm optimization (CSO) is a novel meta-heuristic for evolutionary optimization algorithms based on swarm intelligence. CSO imitates the behavior of cats through two submodes: seeking and tracing. Previous studies have indicated that CSO algorithms outperform other well-known meta-heuristics, such as genetic algorithms and particle swarm optimization, because of complexity, sometimes the pure CSO takes a long time to converge to reach to optimal solution. For improving the convergence of CSO with better accuracy and less computational time, this study presents an improvement structure of cat swarm optimization (ICSO), capable of improving search efficiency within the problem space under the conditions of a small population size and a few iteration numbers. In this paper, an improved algorithm is presented by mixing two concepts, first concept found in parallel cat swarm optimization (PCSO) method for solving numerical optimization problems. The parallel cat swarm optimization (PCSO) method is an optimization algorithm designed to solve optimization problems Based on cats’ cooperation and competition for improving the convergence of Cat Swarm Optimization,, the second concept found in Average-Inertia Weighted CSO (AICSO) by adding a new parameter to the velocity update equation as an inertia weight and used a new form of the position update equation in the tracing mode of algorithm. The performance of ICSO is sensitive to the control parameters selection. The experimental results show that the proposed algorithm gets higher accuracy than the existing methods and requires less computational time and has much better convergence than pure CSO, and the proposed effective algorithm can provide the optimum block matching in a very short time, finding the best solution in less iteration and suitable for video tracking applications.