Abstract: MRI is a very important technique for detecting brain tumour. In this paper, a new hybrid technique is used for the classification of MRI images. This is based on the support vector machine (SVM) and fuzzy c-means for brain tumor classification. This algorithm is a combination of support vector machine (SVM) and fuzzy c-means, a hybrid technique for predicting brain tumor. First the image is enhanced by using enhancement techniques like contrast improvement, and mid-range stretch. Then double thresholding and morphological operations are being used for skull striping. Fuzzy c-means (FCM) clustering, a very important technique of segmentation is used for the image to detect the suspicious region in brain MRI image. Wavelet based Grey level run length matrix (GLRLM) is used for extraction of features from the given brain image, then SVM technique is applied for classifying the brain MRI images, which gives accurate and effective result for the classification of brain MRI images.

Keywords: Support vector machine (SVM), Fuzzy C-means (FCM), MRI, GLRLM, Clustering, Segmentation, Feature Extraction, Enhancement, Skull stripping etc.