Objective Quality Assessment of Image Enhancement Methods in Digital Mammography - A Comparative Study

摘要

Mammography is the primary and most reliable technique for detection of breast cancer. Mammogramsare examined for the presence of malignant masses and indirect signs of malignancy such as microcalcifications, architectural distortion and bilateral asymmetry. However, Mammograms are X-ray imagestaken with low radiation dosage which results in low contrast, noisy images. Also, malignancies in densebreast are difficult to detect due to opaque uniform background in mammograms. Hence, techniques forimproving visual screening of mammograms are essential. Image enhancement techniques are used toimprove the visual quality of the images. This paper presents the comparative study of different preprocessingtechniques used for enhancement of mammograms in mini-MIAS data base. Performance of theimage enhancement techniques is evaluated using objective image quality assessment techniques. Theyinclude simple statistical error metrics like PSNR and human visual system (HVS) feature based metricssuch as SSIM, NCC, UIQI, and Discrete Entropy.