PERFORMANCE COMPARISON OF NOISE REDUCTION IN MAMMOGRAM IMAGES

S. Kannan, N. P. Subiramaniyam, A. T. Rajamanickam, A. Balamurugan

Abstract: Noise level present in mammogram images highly affects the image analysis and accuracy of classification. Hence removing noise present in mammogram images is an important task. Noise present in the medical images depends on imaging modalities. The dominant noise present in mammogram images are quantum noise. The objective of this work is to study the various filters such as mean, median and wiener filter with different window size using standard benchmark (Digital Database for Screening Mammography) DDSM dataset. Higher the value of the (Peak Signal to Noise Ratio) PSNR, better the image quality of the restored image. The restored image quality of various filters was evaluated with PSNR value. We obtained, the wiener filter with window size 3x3 gives better result for noise reduction in mammogram images.

Keywords: Mammogram, Quantum Noise, Mean Filter, Median Filter, Wiener Filter, DDSM and PSNR

DOI: https://doi.org/10.15623/ijret.2016.0502006