在傳輸一影像時，受限於傳輸頻寬的限制，當傳輸品質非常差的時候，因為接收端的門檻效應而使得對影像造成突波雜訊，使得接收者對於影像的辨識度變差。傳統上我們使用中值濾波器來消除這種雜訊，然而此種方法會對影像的細節造成傷害。因此以切換方式為概念的排序統計型濾波器被廣泛的使用來改善去除雜訊時對影像所造成的破壞；其特點乃是利用周圍圖素的特徵，找出最相近原始圖素的數值來取代被污染的圖素同時又盡量保有影像的細節。適用的方法有SD-ROM[3]、SM[18]、New Efficient[31]、Fuzzy filter[32]與MMEM[10]等…濾波器。
所提出的方法乃是以New Efficient[31]這篇為架構，加入模糊理論的概念，目的就是要改進原作者的方法在遇到雜訊圖素跟周圍圖素差異不大時，在門檻值選擇上的困難處，以及在處理影像邊緣與細線條時容易產生因門檻值選擇不當而造成判斷錯誤的情況，進而造成還原回來的效果變差。我們透過模糊法則來描述雜訊的存在程度來改善，並且參考MMEM[10]的方法對影像灰階值為0跟255時的雜訊做改進，最後結合在一起以得到較原先作者對於處理突波雜訊更高之效能。實驗結果圖形可以看到對於影像細部的改進之處，以及PNSR與COR數據上的改善。

In transmission an image in channel, due to a limited of bandwidth, if the channel is so poor that the noise variance is high enough to make the threshold operation done at the front end of the receiver will contribute to impulse noise on images. Traditionally, median filter are used to suppressing this kind of noise but induced the damage on image details. An ordered statistics filter based on switching scheme is popularly used to improve the damage on image detail when we are de-noising; the central concept of the ordered statistics filter based on switching scheme is by use the characteristic of surrounding pixels to find out the most similar value for replace the noisy pixel. The SD-ROM[3], SM[18], New Efficient[31], Fuzzy filter[32] and MMEM[10]…etc.
The proposed method is based on the structure of New Efficient[31], also including the fuzzy concept in order to improve the difficult to get threshold when the difference of noisy pixel and surrounding pixels is not large, and an error judgment may occurred because of the misfit threshold when processing image details and thin lines. Theses situations may cause the poor performance on reducing images. By judgment the existence of impulsive noise is expressed by fuzzy rules for improvement, and referenced the MMEM filter[10] for improve the salt and pepper noise. By combine these methods to get better performance than the New Efficient algorithm. Simulation results can show the improving part in images and also on PSNR and COR value.

[1] C. Gonzalez, E. Woods, “Digital Image Processing”, Second Edition, University of
Tennessee. 2002 Interactive Prentice Hall.
[2] S. Meher, “Development of Some Novel Nonlinear and Adaptive Digital Image Filters
for Efficient Noise Suppression”, National Institute of Technology, Orissa India, 2004.
[3] E. Abreu , M. Lightstone, S. K. Mitra and K. Arakawa “A New Efficient Approach for
the Removal of Impulse Noise from Highly Corrupted Images”, IEEE Trans. on Image
Processing, Vol. 5, No. 6, pp.1012-1025 ,1996
[4] E Abreu, SK Mitra, "Signal-dependent rank-ordered mean (SD-ROM) filters for
removal of impulses from highly corrupted images," Proc. IEEE International Conference
on Acoustics, Speech & Signal Processing, Detroit, MI, May 1995, pp. 2371-2374 .
[5] M Lightstone, E Abreu, SK Mitra, K Arakawa , "State-conditioned rank-ordered
filtering for removing impulse noise in images," Proc. IEEE International Symposium on
Circuits & Systems, Seattle, WA, May 1995, pp. 957-960.
[6] TH Yu and SK Mitra, “Detail-preserving impulse noise removal of images using
modified dynamic programming,” in Proc. /990 mt. Conf Acoust., Speech Signal
Processing, 1990, vol. 4, pp. 1881-1884.
[7] Y Wang, SK Mitra – Acoustics,” Edge preserving image coding based on local
modeling of images”, Acoustics, Speech, and Signal Processing, 1988. ICASSP, pp.
1320-1323 vol.2.
[8] SK Mitra, TH Yu, and R. Ali, “Efficient detail-preserving method of impulse noise
removal from highly corrupted images,” Image Video Process. (SPIE), vol. 2182, pp.
43–48, 1994.
[9] N Strobel, SK Mitra, "Quadratic filters for image contrast enhancement," Proc. 28th
57
Annual Asilomar Conference on Signals, Systems, and Computers, Pacific Grove, CA,
November 1994, pp. 208-212 (with N. Strobel).
[10] W-Y. Han and J-C. Lin, “Minimum-maximum exclusive mean (MMEM) filter to
remover impulse noise from highly corrupted images”, IEE, Electronics Letters, vol. 33,
No. 2, pp 124-125, Jan 1997.
[11] M. Ma, X. Jiao and X. Tan, “Fuzzy Hybrid filter for Removal of Impulsive Noise
from Highly Corrupted Images”, Proceeding of IEEE Conf. TENCON”02 pp.885-888.
[12] V. Crnojevic, “Impulse noise filter with adaptive MAD-based threshold”, ICIP 2005,
Volume: 3, On page(s): III- 337-40
[13] V. Crnojevic, V. Senk and Z. Trpovski, “Advanced Impulse Detection Based on
Pixel-Wise MAD”, IEEE Signal Proc. Lett., vol.-11, No.-7, pp. 589-592, July 2004.
[14] N. C. Gallagher Jr and G.W.Wise, “A theoretical analysis of the properties of median
filters,” IEEE Trans. Acoust., Speech, Signal Processing, vol. ASSP-29, pp. 1136–1141,
Dec. 1981.
[15] T. S. Huang, G. J.Yang, and G.Y. Tang, “A fast two-dimensional median filtering
algorithm,” IEEE Trans. Acoust., Speech, Signal Processing, vol. ASSP-27, pp. 13–18, Feb.
1979.
[16] P. Huber, “Robust Statistics". New York: Wiley, 1981.
[17] Rousseeuw PJ, Croux C. “Alternatives to the median absolute deviation”. Journal of
American Statistical Association, 1993, Vol. 88, pp. 1273-1283.
[18] S. Zhang and M. A. Karim, “A new Impulse Detector for Switching Median Filters”,
IEEE Signal Processing Letters, vol. 9, no.11, pp.360-363, Nov 2002.
[19] T. Chen, K.-K. Ma, and L.-H. Chen, “Tri-state median filter for image de-noising,”
IEEE Trans. Image Processing, vol. 8, pp. 1834–1838, Dec. 1999.
[20] Z. Wang and D. Zhang, “Progressive switching median filter for the removal of
58
impulse noise from highly corrupted images,” IEEE Trans. CircuitsSyst. II, vol. 46, pp.
78–80, Jan. 1999.
[21] Russo F., and Ramponi G., “A Fuzzy Filter for Images corrupted by Impulse Noise”,
IEEE Signal Processing Letters, vol-3, no. 6, pp. 168-170, June 1996.
[22] F. Russo and G. Ramponi, "A Fuzzy Operator for the Enhancement of Blurred and
Noisy Images", IEEE Trans.on Image Processing, vol.4, no.8, August, 1995.
[23] OH Jinsung, LEE Changhoon, KIM Younam, “Minimum-Maximum Exclusive
Weighted-Mean Filter with Adaptive Window” Transactions on Fundamentals of
Electronics, Communications and Computer Sciences 2005 E88-A(9):2451-2454.
[24] S. J. Ko, Y.H. Lee, “Centre weighted Median Filters and their Applications to Image
Enhancement”, IEEE Trans. on Circuits and Systems, vol. 38, pp. 984-983, September
1991.
[25] H. G. Longotham and A.C. Bovik, “Theory of Order Statistics Filters and Their
Relationship to Linear FIR Filters”, IEEE Trans. on Acoust. Speech, Signal Processing, vol.
ASSP-37, no.2, pp. 275-287, February 1989.
[26] Y Dong, RH Chan, S Xu, “A Detection Statistic for Random-Valued Impulse Noise”,
Image Processing, IEEE Transactions on, 2007, Volume: 16, Issue: 4 On page(s):
1112-1120.
[27] A. Kundu. SK Mitra, and pp Vaidyanathan, “Application of two- dimensional
generalized mean filtering for removal of impulse noises from images,” IEEE Trans.
Acoust., Speech, Signal Processing, vol. ASSP-32, pp. 600–609, June 1984.
[28] L. A. Zadeh, “Fuzzy Sets”, Information and control, vol.8, pp. 338-353, 1965.
[29] J. W. Tukey, “Experimental Data Analysis”, Reading, M. A: Addison- Wesley; 1971.
[30] L. Naaman and A. C. Bovik, “Optimal order statistics filters with coefficient
censoring” in Proc. IEEE Int. Conf. Accoust. , Speech, Signal Processing, New York, Apr.
59
11-14, 1988.
[31] Wenbin LUO, “A New Efficient Impulse Detection Algorithm for the Removal of
Impulse Noise” in IEICE TRANS. FUNDAMENTALS, VOL.E88–A, NO.10 OCTOBER
2005
[32] Stefan Schulte∗,etc, “Fuzzy random impulse noise reduction method”, Fuzzy Sets and
Systems, Elsevier Science, 158 (2007) 270–283.
[33] M. E. Yüksel and E. Bes﹐dok, “A simple neuro-fuzzy impulse detector for efficient
blur reduction of imuplse noise removal operators for digital images,” IEEE Trans. Fuzzy
Syst., vol. 12, no. 6, pp. 854–865, Dec. 2004.
[34] Kaoru, Arakawa, “Median filter based on fuzzy rules and its application to image
restoration”, Fuzzy Sets and Systems, Elsevier Science, 77 (1996) 1-13.
[35] Igor Aizenerg, C. butakoff, “Effective Impulse Detector Based on Rank-Order
Criteria”, IEEE SIGANL PROCESSING LETTERS, VOL. 11, NO. 3, MARCH 2004.