Edge Detection in Digital Images Using Ant Colony Optimization (ACO) Algorithm

Abstract

 Human life is greatly influenced by images, which are essential in communication, medical science, defense and industrial applications. The rapid growth of technology has enabled advanced image processing techniques that assist in object recognition, segmentation and feature extraction. Edge detection, being the fundamental step in image analysis, directly impacts the accuracy of higher-level processing tasks. Traditional techniques such as Sobel, Prewitt, Roberts and Canny often suffer from noise sensitivity, high computational complexity and incomplete edge localization. Ant Colony Optimization (ACO), inspired by the collective foraging behavior of ants, has emerged as an effective meta-heuristic for solving optimization problems. In edge detection, ACO utilizes pheromone matrices and heuristic information to efficiently detect edges while minimizing false positives and negatives. This research applies ACO-based edge detection on different sample images (tomato, butterfly, lotus, Taj Mahal, etc.), comparing performance through mean square error (MSE) and peak signal-to-noise ratio (PSNR). Results demonstrate that ACO achieves superior edge localization, robustness to noise and reduced computational complexity compared to conventional methods.