Image Segmentation Method Based Upon Otsu ACO Algorithm

Image segmentation is a very important research content in the fields of computer vision and pattern recognition. As the basis for image understanding and image analysis it is always receives high attention. At present in those commonly used segmentation methods of contrast ratio, margin and grayscale detection the threshold processing is one of the most effective. The threshold methods can also be divided into Otsu, minimum error thresholding, optimum histogram entropy and minimum cross entropy, etc. By combining the advantages of ACO (Ant Colony Optimization) the present paper has designed an ACO segmentation algorithm of solving extra-class variance maximum value to determine the optimum threshold value. The algorithm can quickly and steadily find the optimum segmentation threshold in a non-linear way so as to effectively segment the target and its background, and receive a best result in image segmenting.