Fuzzy logic and distance measure based adaptive fixed value impulse noise filter
- 20 Downloads
The proposed Fuzzy Logic and Distance Measure based Adaptive Filter (FDMA) uses the principles of fuzzy logic and distance measure for the detection and correction of fixed-value impulse noise, respectively. This filter, in the noise detection phase, classifies the pixel of the corrupted image either as uncorrupted or corrupted ones, using fuzzy logic and statistical measures and thereby constructs a flag matrix. In the noise correction phase, the intensity of each corrupted pixel is replaced by an estimated intensity value, computed using its uncorrupted neighbouring pixel, lying within Minimum Euclidian Distance. The noise suppression capability of this filter is quantitatively measured using Peak Signal to Noise Ratio, Mean Structural Similarity Index Matrix and Coefficient of Correlation values. Additionally, the merit of this filter is endorsed by Human Visual Perception of the restored images. The noise restoration ability of FDMA is confirmed to outperform the reported filters. This filter can seamlessly be employed in the application domain wherein fixed-value impulse noise is the primary cause for quality degradation.
KeywordsNoise detection Noise correction Fuzzy systems Euclidian distance Fixed-value impulse noise
Mathematics Subject Classification68U10 94A08
The Authors wish to place on record the financial assistance received in the form of a Major Research Project from UGC, New Delhi. The authors also thank the authorities of GRI for their support.
- 1.Sonka, Milan, Vaclav Hlavac, and Roger Boyle. 2008. Digital image processing and computer vision. Cengage Learning.Google Scholar
- 2.Gonzalez, R.C., and R.E. Woods. 2009. Digital Image Processing, 3rd ed. Upper Saddle River: Pearson Prentice Hall.Google Scholar
- 3.Edward R. Dougherty, and Jakko T. Astola. 1999. Nonlinear Filters for Image Processing, IEEE & SPIE Press.Google Scholar
- 6.Satpathy, S.K., S. Panda, K.K. Nagwanshi, and C. Ardil. 2010. Image restoration in non-linearing domain using MDB approach. IJICE Journal 6 (1): 45–49.Google Scholar
- 7.Forouzan, Amir R., and Babak Nadjar Araabi. 2003. Iterative median filtering for restoration of images with impulsive noise. Proceedings of ICECS 1: 232–235.Google Scholar
- 8.Abreu, Eduardo, and Sanjit K. Mitra. 1995. A signal-dependent rank ordered mean (SD-ROM) filter-a new approach for removal of impulses from highly corrupted images. Proceedings of the ICASSP 4: 2371–2374.Google Scholar
- 12.Shanmugavadivu, P., and Eliahim Jeevaraj, P. S. 2011. Fixed—value impulse noise suppression for images using PDE based adaptive two-stage median filter. In ICCCET-11 (IEEE Explore), pp. 290–295.Google Scholar
- 13.Shanmugavadivu, P,. and Eliahim Jeevaraj, P. S. 2012. Laplace equation-based adaptive median filter for highly corrupted images. In International Conference on Computer Communication and Informatics (ICCCI-2012), pp. 47–51.Google Scholar
- 14.Shanmugavadivu, P., and Eliahim Jeevaraj, P. S. 2011. Fixed-value impulse noise suppression for images using PDE based adaptive two stage median filter. In International Conference Computer, Communication and Electrical Technology, pp. 290–295.Google Scholar
- 16.Zhong-gui, Sun, Liaocheng Chen Jie, Meng Guang-wu. 2008. An impulse noise image filter using fuzzy sets. In International Symposiums on Information Processing (ISIP), 2008, pp. 183–186.Google Scholar