Modified Ranked Order Adaptive Median Filter for Impulse Noise Removal
In this paper problem of impulse noise removal is considered. Specifically, modifications of ranked order adaptive median filter (RAMF) are proposed. RAMF is popular, well established and effective switching median filter for denoising images corrupted by impulse noise. However, the modifications proposed in this paper significantly improve its results, especially in case of highly corrupted images. Results of denosing of images under a wide range of noise corruption (5-95%) using the original and the modified ranked order median filter are presented, compared and discussed. Comparison is made by means of PSNR and SSIM index.
KeywordsImpulse Noise Noise Removal Noise Detection Noise Density Noisy Pixel
Unable to display preview. Download preview PDF.
- 1.Arce, G.R., Gallagher, N.C., Nodes, T.: Median filters: theory and applications. In: Huang, T. (ed.) Advances in Computer Vision and Image Processing. JAI Press, CT (1986)Google Scholar
- 2.Astola, J., Kuosmanen, P.: Fundamentals of nonlinear digital filtering. CRC Press, USA (1997)Google Scholar
- 5.Gonzalez, R.C., Woods, R.E.: Digital Image Processing. Prentice Hall, USA (2007)Google Scholar
- 11.Vijaykumar, V.R., Vanathi, P.T., Kanagasabapathy, P., Ebenezer, D.: High density impulse noise removal using robust estimation based filter. IAENG International Journal of Computer Science 35(3) (2008), http://www.iaeng.org/IJCS/issues_v35/issue_3/index.html (accessed November 2010)
- 15.Zvonarev, P.S., Apalkov, I.V., Khryashchev, V.V., Reznikova, I.V.: Neural network adaptive switching median filter for the restoration of impulse noise corrupted images. In: Kamel, M.S., Campilho, A.C. (eds.) ICIAR 2005. LNCS, vol. 3656, pp. 223–230. Springer, Heidelberg (2005)CrossRefGoogle Scholar