Designing edge detection filters using Particle Swarm Optimization

This paper presents a novel edge detection method based on Particle Swarm Optimization. Unlike classical filters that are set by intuitive knowledge, a new filter is proposed on the basis of evolutionary computation. A proper synthetic training image and its edge map are used to find an optimum edge filter. The advantage of this method is that an effective edge detection filter can be easily constructed. Provided results certify that our proposed method outperforms commonly used edge detection algorithms.