Image Edge Detection Method Based On Ant Colony Algorithm Is Studied

Image edge detection is a technique for recognizing the field of digital image and analyzing digital images, and even studying the depth field of digital image. In the mid1950s people who inspired by the natural behavior propose some optimization problems. The new algorithms are created to solve combinatorial optimization problems, such as ant colony algorithm, artificial neural network algorithm, genetic algorithm, as well as the integration of several algorithms. Image is often considered as matrix by the each pixel position of the image. So image edge detection can be attributed to the optimization problem for solving the discrete combinations. The swarm intelligence of ants has advantages in itself, while is positive feedback and robust distributed. It is good for deal with optimization problem, and some kind of problem can convert into combinatorial optimization.The main content of this dissertation is as follows:Firstly, this dissertation introduces the background and significance of this subject research, as well as analyze the status of edge detection algorithm.Secondly, we illustrate the foraging behavior of ant colony, explain the basic idea of the ant colony algorithm.And describe the characteristics of ant colony algorithm.The proposed approach exploits a number of ants, and establish a pheromone matrix, which represents the edge information at each pixel location of the image.Thirdly, we research and analyze the implementation of ant edge detection algorithm. The robust can be applied with only minimal change to other combinational optimization problems. Experiments have found that plus pheromone matrix of four heuristic functions. Describe the effect of ant colony edge detection method, and it is good for the noise image.Finally, it is complexity and randomness for us to select the parameter of ant colony edge. So the last chapter of this paper is to discuss how to use the genetic algorithm to select the parameters of the ant colony algorithm. Using genetic algorithm to choose parameter can avoid uncertainties and blindness.A comprehensive summary of the work done in this paper. Put forward the shortcomings, and predicts the content of this paper will continue to study.