Among The Various Types Of Human Cancers, Breast Cancer Is The Most Common Form, Among The Women. However, Complete Curing Of The Said Disease Is Possible If It Is Detected In Its Early Stage. Early Detection Of Breast Cancer Improves Survival Rate Of Women. Mammography Is Currently The Method Of Choice For Early Detection Of Breast Cancer In Women. However, The Interpretation Of Mammograms Is Largely Based On The Radiologist’s Opinion. The Radiologist’s Performance Increases When They Incorporate Automatic Image Analysis In Their Decision Making Process For Both The Detection And Diagnosis Of Cancer. This Paper Proposes A Novel Approach For The Development Of A Computer Aided Decision System To Automatically Detect Abnormalities In Mammograms. Mathematical Morphology Proves To Be A Useful Tool For The Detection Of Abnormalities In Digital Mammograms. We Proposed A New Algorithm For The Detection Of Abnormalities On Mammograms. Every Suspicious Object Is Marked Using A Binary Image, Which Is Used As A Mask For Object Extraction From The Original Image. The Features Of The Extracted Objects Are Classified Using Naive Bayes Classifier.
