Comparative Study Of Different Image Denoising Filters For Mammogram images

Abstract

Since many years, breast-cancer is the major reason for the deaths related to cancer amongst women. It also affects the women’s physical and mental health. If it is not detected in the beginning then it would increase the ratio of demise. So, its detection at the very first stage is of utmost important. Old techniques are not any more effective. Mammography is counted as a most powerful practice to detect the breast cancer at early stages. And also, it is very useful for the CAD (computer aided diagnosis) technique which is suggested as a second choice by the radiologists. In order to diagnose the breast cancer automatically, it is used. Its early detection depends on two things: one is quality of mammography images and the other is capacity of the radiologists to read it . These pictures of mammography are a bit complex and will contain a lot of additional things like noise, artifacts, identificational labels, etc. If it is not removed then it will affect the other stages of the observation of breast cancer. So, the pre-processing of the mammography images is of utmost important before moving to the further steps for the observation of breast cancer. There are various steps in diagnosis of breast cancer like: segmentation, feature extraction, feature selection and classification. But in this report we will discuss about the most important various pre-processing techniques and various segmentation techniques used for the removal various additional things present in the mammography images. It will give a brief idea on the various trending and effective pre-processing and segmentation techniques.