EDGE DETECTION IN DIGITAL IMAGES USING EVOLUTIONARY ALGORITHM

Abstract

An edge is an abrupt change in intensity that indicates the boundary between two adjacent regions. Its detection is the fundamental operation in image processing. Evolutionary algorithms can be used for efficient results. Evolutionary algorithms are a set of algorithms that are inspired from nature. Various algorithms have been developed in the past few years. Artificial Bee Colony algorithm is one such algorithm that copies the behavior of bees in the nature. It is used for detection of edges in the gray scale digital images. It uses the concept of 3 types of bees in the bee colony – Employee Bees that searches for the food source in the search space and inform about it to the colony, Onlooker Bees that selects one location among the previously informed locations based on the fitness values and search for food source in its neighborhood and Scout Bees move to random locations. The proposed algorithm results in edge-map that contains no thick edges. Also the corners are not missed. It is faster than edge detectors that use Particle Swarm Optimization algorithm, Ant Colony Optimization algorithm or Gravitational Search algorithm. The algorithm is then tested on different sample images and results have been compared to the results of the most commonly used edge detectors. The technique can be enhanced further for detection of edges in colored images.