Abstract
Breast cancer is one of the prominent causes of female mortality in the world, and microcalcification clusters are the important indicators for breast cancer. Mammography is a useful procedure for revealing these indicators at an early stage. But the manual interpretation of microcalcifications is difficult due to low contrast with the background parenchymal tissue. This makes it hard to judge the size, shape and morphology of the microcalcifications. In this paper a methodology, which is a combination of morphological operations, unsharp masking and Gaussian filter, has been proposed for enhancement of mammograms to bring out the tiny details of microcalcifications present in a variety of nonhomogeneous background tissues while restoring their shape and size. For experiment the mammogram images, collected from Digital Database for Screening Mammography, have been used and the results are compared to standard methods like contrast limited adaptive histogram equalization, multi scale top-hat transform based algorithm and bi-histogram equalization with adaptive sigmoid functions. The results from both the qualitative and quantitative evaluations suggest that the proposed methodology is very effective.
http://ift.tt/2sSd4OA
Δεν υπάρχουν σχόλια:
Δημοσίευση σχολίου