Comparison of Feature Extraction Techniques to classify Oral Cancers using Image Processing

Abstract— Oral cancer is the sixth most vulnerable cancer in the world, which accounts for over thirty percent of all cancers reported in the country and oral cancer control is quickly becoming a global health priority. In this work, a system is developed to segment, extract features and classify cancers. Later, a comparison is made. The proposed system consists of fve steps. First, the images are enhanced and the Region of Interest (ROI) is segmented using Marker Controlled Watershed Segmentation. Feature Extraction methods like Gray Level Co-occurrence Matrix (GLCM), Intensity Histogram and Gray Level Run Length Matrix (GLRLM) are used to extract features from ROI. Next, classification is made using Support Vector Machine (SVM) classifier to classify the tumor as benign or malignant mass and a comparative study is performed to identify the best feature extraction technique. Cancer has become one of the leading causes of death in India. Oral cancer has high mortality ratio among all malignancies. It constitutes 17% of all cancers in males and 10.5% of all cancers in females making it the commonest cancers in males and the third commonest cancer among females. < final year projects >