Binarization based edge detection using universal law of gravity and ant colony optimization

Abstract

Swarm Intelligence is a relatively new approach for solving complex computational problems. An ant colony optimization (ACO) is based on the natural behavior of ants which deposit pheromone on the ground during their trails for foraging. ACO exploits the sensing capabilities of the group and avoids the premature convergence by way of distributed computing. This paper presents a new approach for edge detection using a combination of binarization, ACO and universal law of gravity. In the proposed approach, edge detection problem is handled by first converting the input gray scale image into binary image so that the image matrix consists of only two gray levels, i.e. 0 and 255. Ants are then placed randomly all over the image which would then undergo foraging in search for food (edge pixels). The proposed approach uses the law of universal gravity to calculate the heuristic function which acts as the way to food source for the artificial ants to detect the edge pixels.