Screening and early detection of abnormalities needs an automated system that identifies the abnormalities and cancer in the MRI images as early as possible. An automated system that segments the abnormalities and identifies the defect in the MRI images is proposed. The cancer in the MRI images were identified by manual labeling of the tumor portions in the images. The tumor portion in the images were identified and also the state of the tumor is also identified. The method identification of the tumor is based on the manual labelling of the tumor portions. The tumorous portions were recognized with the help of the identification of the feature points in the images using mean value calculation. The images were divided into small patches. The patch separation process takes place by identifying the seed point around the image. The image pixels around identified seed points were then taken. The features were extracted from the images by calculating statistical informations from the image such as mean and standard deviation. The statistical based informations were much helpful to identify the portions in the image uniquely. The texture based informations were also needed to identify the tumorous portions more clearly. Gabor filters possess the optimal localization properties in both spatial and frequency domain. The impulse response of these filters is created by multiplying an Gaussian envelope function with a complex oscillation The phase-difference of the left and right filter responses to estimate the disparity in the stereo images. The identified patches around the seed points were then classified into beningn stage or malignant stage. The stage of tumor is identified using SVM classifier. Different type of SVM such as PSVM and FSVM were implemented and the performance is compared with different methodologies by comparing the Accuracy, Sensitivity, and Specificity. The classification process employed using SVM produces better results compared to the other segmentation methods and the process is capable of the identification of different types of tumors differently. The different type of tumors were identified and they were denoted in separate manner.