Digital Mammography has become the most effective technique for detecting the early stages of breast cancer. Conventional denoising methods such as mean, median, Gaussian, wiener, and adaptive median filters are discussed in this paper to enhance the image quality. For testing scenarios, the Mammographic Image Analysis Society (MIAS) database is used to carry out the estimation of performance parameters. By comparing the different techniques concerning performance parameters the best method for denoising mammogram images is to be determined. In the further process implementation of thresholding and Segmentation are used to mine the suspicious sections from the DM. The mined section is then evaluated using the Gray level co-occurrence matrix (GLCM) to know the severity of the disease by examining its texture features. Later in this paper, using the above texture features DM is classified as benign and malignant cells. Classification is carried out using a classification learner app with various machine learning algorithms.
Software And Hardware