The Application Of Support Vector Machine In Image Segmentation Based On Edge Detection And Function Regression

Support Vector Machine (SVM) is a new method of machine learning. It bases on the Statistical Learning Theory, and can settle "small" example problem well. Because of its excellent learning ability, SVM has been applied to such fields as Text Classification, Face Recognition, and Figure Recognition. In this paper, we will discuss the application of SVM in image segmentation based on edge detection and function regression.In image segmentation, we use support vector classification (SVC) to classify the gray value of pixels, which belongs to the image. The so-called " support vectors" decide the edge of image, so the segmentation can be completed. In the paper, two methods of SVC are used. One is binary-class the other is single-class (one-class). The former belongs to supervised learning and the latter belongs to unsupervised learning. The distinct difference between supervised learning and unsupervised learning lies in whether the example consists of the pre-processed output value. The detail description is in the third chapter.In this paper, we deal with two problems of function regression. They are estimating parameter and boundary problems of differential equations. When solving the problems, we use the support vector regression (SVR). First assuming the formula of function, then according to the differential and boundary conditions we transform the original problem to the quadratic programming problem. Finally, use the learning algorithm of SVR to decide the parameters.