Classification of Mammograms using Texture Features

Abstract

Breast cancer is the leading cause of death among women aged 35 to 54 which gives the need of prevention of breast cancer at an early stage. Mammography is the process of detecting and screening breast cancer at an early stage and prevents death. The mammogram image has to be processed to extract the features in the image. Feature extraction is done based on the Gray-level co-occurrence matrix (GLCM). The mammogram image is the input and the classified image is the output which categorizes the image with the categories namely normal, benign and cancer.  Classification is done using k-nearest neighbor classifier. The classifier calculates the distance between the query image and the images in the database and assigns class name to the query image for which the distance is least, as the output.