Wavelet Based MRI Brain Image Classification Using Radial Basis Function in SVM

This paper presents a novel method to classify MRI brain images as normal or abnormal image using Radial Basis Function (RBF) in Support Vector Machine (SVM). This method first extracts wavelet derived features, followed by dimensionality reduction of features using Principle Component Analysis (PCA). The resultant features are given as input to SVM that uses RBF, a non-parametric and flexible model, for classifying images. To enhance generalization of this kernel SVM the strategy of K-fold stratified cross validation is used. The classification results are compared with linear and polygonal kernels, and found that the RBF kernel achieves the highest classification accuracy as 95.25%.