A brain tumor is formed when abnormal cells get accumulated within the brain. These cells multiply in an uncontrolled manner and damage the brain tissues. Magnetic Resonance Image scans are commonly used to diagnose brain tumors. However, segmenting and detecting the brain tumor manually is a tedious task for the radiologists. Hence, there is a need for automatic systems which yield accurate results. A fully automatic method is introduced to detect brain tumors. It consists of five stages Image Acquisition, Preprocessing, Segmentation, using Fuzzy C-means technique; Harris Corner Detection based feature extraction and classification using K-NN. Performance metrics such as accuracy, precision, sensitivity and specificity are used to evaluate the performance.