HISTOGRAM NORMALIZATION TECHNIQUE FOR PREPROCESSING OF DIGITAL MAMMOGRAPHIC IMAGES

Abstract

Digital mammogram has become the most efficient tool for early breast cancer detection modalities and pre-processing these images requires high computational capabilities. Pre-processing is one of the most important step in the mammogram analysis due to poor captured mammographic image qualities. Pre-processing is basically used to correct and adjust the mammogram image for further study and classification.  Many image pre-processing techniques have been developed over the past decades to help radiologists in diagnosing breast cancer. Most studies conducted have proven that a pre-processed image is easier for radiologist to accurately detect breast cancer especially for dense breast. Different types of techniques are available for pre-processing of mammograms, which are used to improve image quality, remove noise, adjust contrast, enhance the image and preserve the edges within the image. This paper acquired 20 digital mammograms from Mammographic Image Analysis Society (MIAS) database and uses Histogram Normalization algorithm for pre-processing of the mammograms. A percentage of 95% was obtained. It was observed that the pre-processed mammographic images displayed breast abnormalities clearer with little or no noise.