Novel Approach Of Edge Detection In Mammographic Images

Edge Detection leads to reducing the search space and time required for processing. There are many existing edge detection techniques but still many practical issue exists in this field. Edge is a set of those pixels whose grey have step change/ rooftop change and it exists between objects & its background, onject & object, region & region and between element & element. These edge detection techniques do not work well upon these images. Many edge detection algorithms are used to find boundaries in images. A significant problem with mammography is the variability of diagnosis among radiologist and computer aided diagnosis (CAD) methods that consistently highlight region on mammograms that may warn further examinations. One of the first step in CAD is the segmentation of the image into background and the breast. This has the advantage of further simplifying the processing of image (by eleminating the background) and also provides as a reference for the alignment of the views when two viewes are being compared. Knowledge of the breast edge also sometimes identifies large masses that have distorted the outline of the breast.