Cumulative Histogram based Dynamic Particle Swarm Optimization Algorithm for Image Segmentation

Abstract

Image Segmentation is an essential part of image processing to recognize an image.  It is the process to find the subsets of an image on the basis of its characteristics such that values or/and position of pixels. The computational complexity of these type of problems are considered as an NP-complete problem, it is a very hard task to find the solution to such types of problems. A heuristic algorithm like GA, ACO and PSO can be used to find the solution of such type of problems.  In the proposed approach the Dynamic Particle Swarm Optimization (DPSO) was used to find the approximate solutions for Image segmentation.  Cumulative Histogram will be used to find initial random populations of the Swarm. The result found from this algorithm is tested at different levels of measurement and compare with existing PSO algorithm