A new homogeneity-based approach to edge detection using PSO

In the ideal case, the result of applying an edge detector to an image may lead to a set of connected curves that indicate the boundaries of objects, the boundaries of surface markings, as well as curves that correspond to discontinuities in surface orientation. Many edge detection algorithms act as a convolution matrix and are applied just on one pixel and its neighbors like Sobel and Kirish operators, whereas edges on an image are a collection of pixels which are recognized as an edge. We propose a new edge detector which uses PSO (Particle Swarm Optimization) for detection of best available important curves in an image that represent boundaries of objects. Our main idea is to find best fitting curves on edges of an image based on PSO where particles represent these curves. Also, we introduce two new measures, homogeneity and uniformity factor of a curve, that are used to form the objective function of the PSO based edge detection algorithm. The results show that the proposed algorithm performs better than Sobel and the homogeneity operators and that it can be applied to noisy images without using any filtering algorithms.