Preprocessing filters for mammogram images: A review

Breast cancer is a standout amongst the most widely recognized kind of growth among ladies that develops from breast tissue. Still the exact cause of the breast cancer remain unknown. Early detection and diagnosis is the best and most effective strategy to control the tumor progression. Mammography is the currently recommended imaging method for early determination and diagnosis of breast malignancy. A mammogram can identify abnormal areas in the breast that look like a cancer but turns out to be normal, this leads to false positive. Mammogram images are found to be difficult to interpret so a CAD is becoming an increasingly important tool to assist radiologist in the mammographic lesion interpretation. Preprocessing was considered as an important step in mammogram image analysis. Accuracy of preprocessing will determine the success of the remaining process such as segmentation, classification etc. In this paper, mean, median, adaptive median, Gaussian and wiener denoising filters are used to remove salt and pepper, speckle and gaussian noises from a mammogram image and these filters were compared based on the parameters such as PSNR, MSE and SNR to determine which filter is better for removing these noises in mammogram images.