Adaptive edge detection using ant colony

Abstract

In this paper, an adaptive edges detection method based on ant colony algorithm is presented. Ant colony algorithm is a swarm-based metaheuristic inspired by the self-organizing properties of ant colony in nature. Artificial ants in movement create a pheromone graph, which denotes data of edge image. Further behaviors were added to each ant in response to local stimuli: the ant can self-reproduce and lead its progenitors in an appropriate direction to enhance research in suitable areas and it can die too if it exceeds a specific age and so eliminate the ineffective search. Experimental results show the performance of this technique enriched with these behaviors. It provides a good segmentation, fast and adaptive in extracting edges for a variety of images.