Detection of Brain Tumor Using Combination of BWT and K-SVM

Abstract

In this paper, a novel brain tumor detection technique has been presented with combined BerkelyWavelet Transform (BWT) and Kernal Support Vector Machines(K-SVM). The BWT based segmentation was presented with k-means clustering to enhance the effective image segmentation. Initially, the wavelet transform was applied to extract features from MRI images using (PCN) for the reductions of features dimensionality. To a K-SVM, the selected features have been given to classify the image. Hence the accuracy and quality rates of image segmentation was improved by this proposed approach.