Research On Image Segmentation Of Maximal Of Two Dimensional Entropy Based On Ant Colony Optimization Edge

Image segmentation is an important part of image processing, is the basic premise for visual image analysis and pattern recognition, and is also the hotspot research at home and abroad in currently. It has important significance for image feature extraction, image analysis and recognition, computer vision, etc. Because images vary widely, making the image segmentation is a classic problem in image processing.This paper analyzes the research status and main problems in image segmentation, focuses on the threshold and edge-based image segmentation method, analysis the significance of the ant colony optimization algorithm applied to the image segmentation, proposed a method based on search edge of ant colony optimization algorithm. Systematically expounded the basic principles of ant colony optimization algorithm, theoretical models, algorithms characteristics, as well as the selection of various parameters. First, using the two-dimensional Gaussian filter of Canny edge detection operator to the image filtering, using the characteristics of search optimization in ant colony optimization algorithm combining Canny edge detection operator to find the optimal solution, to get the edge image for an important source of information in image segmentation. This paper has established a strict sense of segmentation threshold selection method based on the overall information of image. Experimental results show that the method of this paper can work to improve the accuracy of edge detection and image segmentation while preserve the greatest amount of information. Well preserves the image detail information when the image background more complicated, effective balance between global and local information of image, also has a strong adaptive ability in anti-noise, and can lay a good foundation for the next image analysis, feature extraction and image understanding.