Research on image preprocessing and segmentation of medical image

this paper presents a medical image preprocessing based on an adaptive window and the segmentation algorithm of medical image based on support vector machine, using the segmentation algorithm may be the same target which is divided into a plurality of adjacent segments. According to the pattern recognition theory, if a group of training samples can be separated by an optimal hyperplane, upper boundary of the test sample classification error rate expectations are in the training sample support vector average of the total number of training samples proportion, that reduce the influence of subjective factors, which makes segmentation result that has uniqueness.