A Novel Metaheuristic Algorithm for Edge Detection Based on Artificial Bee Colony Technique

Many techniques have been proposed in image edge detection's area, but until today, there is no universal or optimal methods that satisfy all the constraints. Each one had its limitations and its inconvenient. So, in order to create a system that offers a better quality of boundaries detecting in images, we used the Artificial Bee Colony's (ABC) algorithm with Otsu's multilevel thresholding method in different color spaces ABC-Otsu. The performance of the approach is compared with the Ant Colony optimization algorithm (ACO). Berkeley (BSDS500), Oxford-17 Flowers and Drive data-sets were used for experimentation. The theoretical analysis and the experimental results are encouraging and demonstrated that our method outperformed these techniques. Also, the execution time is improved and the obtained results show good qualities too.