Brain Tumor is one of the major threat confronted by many people around the world. As per International Agency of Research on Cancer (IARC) more than one million people are diagnosed with brain tumor per year around the world, with increased fatal rate. During brain tumor studies, the occurrence of the abnormal tissues is easily detectable most of the time, still accurate segmentation and characterization of these abnormalities are not genuine. In the present scenario, the radiologists have to manually study the tumors with the available medical imaging tools and generate a report. The process is time consuming. Although many progresses have been made, but segmentation of brain tumors from MR Images in a quick, accurate, authentic and reproductive way is still a challenging issue. To overcome this problem a system which will detect the tumor and will classify them as benign and malignant has been proposed in this paper by using image processing in integration with machine learning. Which will help to detect the tumor and classify them into benign and malignant in quick time. In this work step by step procedure for image pre-processing, segmenting brain tumor using morphological operations, extracting tumor feature using DWT and classification of the tumor using SVM is accomplished with the actual clinical data. Key word -Brain Tumor, MRI (magnetic resonance imaging), DWT (Discrete wavelet transform), PCA (Principal component analysis), SVM (Support vector machine