Edge detection of digital images using a conducted ant colony optimization and intelligent thresholding

An edge detection algorithm based on Ant Colony Optimization (ACO) and Fuzzy Inference System (FIS) and neural network is presented. This algorithm uses a FIS with 4simple rules to identify the probable edge pixels in 4 main directions, then the ACO is applied for assigning a higherpheromone value for the probable edge pixels rather than otherpixels so that the ants movement toward edge pixels get faster. Another factor that needs to be considered in order to conductthe ants’ movement is the influence of the heuristic information in the movement of any ant to be proportional to local change inintensity of each pixel. Finally, by using an intelligent thresholding technique which is provided by training a neuralnetwork, the edges from the final pheromone matrix areextracted. Experimental results are provided in order to demonstrate the superior performance of the proposed approach

Keywords:

Ant Colony Optimization , Edge detection , Intelligent thresholding