Image segmentation by intelligent clustering technique

Abstract

Image segmentation based on clustering techniques still remains a challenging task. The aim of clustering is to generate homogeneous groups of data. In this paper, we present bio-inspired formulation to perform image segmentation. Specifically, we used the Bird flocking algorithm that uses the concepts of a flock of agents, e.g. birds moving together in a complex manner with simple local rules. Each bird representing one data, move with the aim of creating homogeneous groups of data in a two dimensional environment producing a spatial distribution that can be used to solve a particular computational problem. These characteristics have been used to solve the task of segmentation of images which optimize the partition of image data into homogenous regions.