Skip to Main Content
Due to complexity and irregulation of each encephalic tissue boundary, three-dimensional (3D) reconstruction for MRI image has been a hot area. Support vector machine (SVM) based on statistical learning theory is mainly utilized in classification and regression. One Class SVM (OCSVM) was originally proposed for solving some special classification problems. In this paper, OCSVM, which tries to find the smallest hypersphere enclosing target data in high dimensional space by kernel function, is firstly explored into the application to 3D reconstruction. However, selecting parameters for OCSVM is a complicated problem. In order to reduce the blindness of parameter selection and perfect SVM theory, Immune Algorithm (IA) and K-fold cross validation are introduced to intelligently search optimal parameter. The experimental results demonstrate OCSVM is effective with high reconstruction accuracy.