Segmentation of pectoral muscle in mammograms using fractal method

Abstract

This paper proposes a novel method for the detection and segmentation of pectoral muscle in mammograms using fractal-based analysis. The region of pectoral muscles containing the brightest pixels has an adverse effect, during the detection of microcalcifications in a mammogram and hence detection and segmentation of pectoral muscle is deemed as an important preprocessing step. This proposed technique segments the pectoral muscle using the principle of Fractal dimension of the image, which is image-dependent, in combination with morphological operations. This Fractal technique of pectoral muscle segmentation is proved to produce quite promising results that substantiate the merit of the proposed technique, in addition to computation simplicity.