Intensity-based detection of microcalcification clusters in digital mammograms using fractal dimension

Abstract

This paper presents a novel method to locate and segment the microcalcification clusters in mammogram images, using the principle of fractal dimension. This proposed technique detects the edges using the intensities of the regions/objects in the image, the Fractal dimension of the image, which is image-dependent in such a way that leads to the segmentation of microcalcification clusters in the image. Hence this fractal dimension based detection of microcalcifations is proved to produce excellent results and the location of the detected microcalcifications clusters complies with the specifications of dataset of the mini-MIAS database accurately, which substantiate the merit of the proposed technique.