ABSTRACT

Magnetic Resonance Imaging (MRI) is the most significant approach, in sensing the life threatening illness like brain tumor. During this research data mining techniques are applied for classification of MRI pictures. The proposed contrivance is a fusion of Support Vector Machine (SVM) and fuzzy methods for brain tumor detection. An innovative hybrid approach based on the merger of Support Vector Machine (SVM) and fuzzy c-means is suggested for brain tumor classification. In this process enhancement techniques such as spatial domain method and frequency domain method are used to enhance the picture. Skull striping is performed by mathematical morphology methods and Fuzzy cmeans (FCM) clustering is applied for the segmentation of the picture to expose the incredulous domain in brain MRI pictures. The Gray Level Co-ocurrence Matrix (GLCM) method is used for extracting texture features from the brain pictures, subsequently SVM method is applied to categorize Tumor and Non-Tumor brain MRI pictures, which gives furnish and more efficacious outcome for classification of brain MRI pictures.