A brain tumor is an abnormal growth of cells in the brain which hinders the proper function of the brain and eventually lead to death if not treated properly. The probability of survival rate can be improved if the tumor is detected early and classified properly. Manually detecting and segmenting brain tumors in today‘s brain Magnetic Resonance Imaging (MRI), where a large number of MRI scans taken for each patient, is tedious and subjected to inter and intra observer detection and segmentation variability. Therefore, there is a need for computer aided brain tumor detection and segmentation from brain MR images to overcome the tedium and observer variability involved in the manual segmentation. As result a number of methods have been proposed in recent years to fill this gap, but still there is no commonly accepted automated technique by clinicians to be used in clinical floor due to accuracy and robustness issues. In This thesis we proposed a new segmentation method that can overcame the drawback of global thresholding in identifying regions that has lower intensity, to extract brain tumors in MRI images, which is based on symmetrical side analysis combined with automatic thresholding to extract tumor part from the brain. The anticipated method can effectively be useful to distinguish the shape of the tumor and its geometrical measurement. Also in this thesis, we used a histogram of oriented gradient (HOG) to extract the features and Support Vector Machine (SVM) for brain tumor classification. Based on the confusion matrics result observed the proposed method has produced a better result with F-measure of 0.8031 or 80.31% accuracy than the other two segmenting mechanisms which are global thresholding and Sobel edge detection which has f- measure of 0.685 and 0.7021.