Biomedical Image Edge Detection using an Ant Colony Optimization Based on Artificial Neural Networks

Ant colony optimization (ACO) is the algorithm that has inspired from natural behavior of ants life, which the ants leaved pheromone to search food on the ground. In this paper, ACO is introduced for resolving the edge detection in the biomedical image. Edge detection method based on ACO is able to create a matrix pheromone that shows information of available edge in each location of edge pixel which is created based on the movements of a number of ants on the biomedical image. Moreover, the movements of these ants are created by local fluctuation of biomedical image intensity values. The detected edge biomedical images have low quality rather than detected edge biomedical image resulted of a classic mask and won’t result application of these masks to edge detection biomedical image obtained of ACO. In proposed method, we use artificial neural network with supervised learning along with momentum to improve edge detection based on ACO. The experimental results shows that make use neural network are very effective in edge detection based on ACO.