Edge Detection Using Ant Colony Optimization: A Bio-Inspired Approach For Enhanced Image Segmentation

An essential function of image processing and computer vision, edge detection is fundamental in tasks including image segmentation, object recognition, and scene analysis. Often facing difficulties, including sensitivity to noise, reliance on manual parameter tuning, and the generation of fragmented edge maps, traditional edge detection methods, including the Sobel, Prewitt, and Canny operators, inspired by the natural foraging behavior of ants, this work presents a new edge detection method called Ant Colony Optimization to get beyond these constraints. Using standard test images, the technique was assessed and showed consistent improvements over more traditional approaches. Higher values in signal-to-noise ratio, precision-recall balance, and structural edge continuity are shown by quantitative data. Visually, the technique reduced background interference and produced more connected, finer, cleaner edges. These results verify the efficiency of the Ant Colony Optimization-based method as a strong and exact edge detection system fit for complicated and noise-sensitive image analysis uses.