隨著科技日新月異的進步，數位相機已逐漸地取代傳統的相機，並廣泛的使用在各個領域上：太空探勘、生醫影像、影像檢測、或是使用數位相機記錄日常生活。而影像雜訊的處理，屬於影像處理中的前處理，目標是要保留較多的影像細節，而把不必要的雜訊盡量的抹除，讓後續處理影像資訊的人，可以得到更好的資訊，而不會受到雜訊的汙染的影響，而影響到後續的判斷。

數位影像雜訊濾除的主要架構，分為兩個部分：雜訊的偵測、與雜訊的抹除，
本篇論文使用中值濾波器當作基礎，並結合模糊理論賦予每個像素權重值，並進一步的減少可能誤判的雜訊。在反覆修正的過程中，會使用加權平均的概念，將像素點為雜訊的權重值降低，並提高不為雜訊的權重值，使得有良好的濾波效能。

由最後的實驗結果可以看出，以本研究方法處理固定型脈衝雜訊的效能大幅領先其他濾波器（PSNR成長約2~3dB），而在處理隨機脈衝雜訊的效能雖然沒有領先其他濾波器很多（PSNR成長約1~2dB），但也有不錯的效果。而本篇論文主要的優勢，在於反覆修正的地方只用有可能有雜訊的點加以修正，不像其他篇論文，在疊代的地方使用整張圖形，因此低密度雜訊的狀態下，可以大幅減少運算時間，而又有不錯的濾波品質。

As the technology advances, the digital camera have been gradually taking over the ground of traditional cameras, and are widely in use amidst various domains: space exploration, biomedical imaging, image detection, or simply using a digital camera to record our daily lives. The image noise processing, the pre-processing part of the image processing flow, strives the goals to retain more image details, and to minimize unnecessary noise as less as possible, so that the people doing subsequent image information processing can get better information, and avoid the interference caused by noise pollution, which might affect the subsequent analyzing.

The main architecture of digital image noise filtering can be divided into two parts: the noise detection, and noise erasing. This paper explores the use of a median filter as the basis, combining fuzzy weights assigned to each pixel value, and the reduction of the possibility of false-positive noise. In the iterative correction process, the concept of using the weighted average adjoins, the weight value to be reduced for the noise pixels while the weight value to be increased for normal pixels, in order to obtain a good filtering performance.

The final experimental results indicate that the fixed impulse noise has significant superior performance ahead of other filters, while the performance of random impulse noise also shows a good result, it is not a huge ahead compared with other filters. The main advantage of this thesis is that it proposes to place repeated corrections only on possible noisy points, unlike other researches using the repeated corrections on iterations of entire graphics, so therefore under the state of low noise density, the computation time can be significantly reduced, and a good filtering quality can be delivered.

[1] http://www.techbang.com/posts/1026-bird-column-great-really-better-on-the-light-sensitive-digital-camera-device-a
[2] R. C. Gonzalez and R. E. Woods, Digital Image Processing, Third Edition. Pearson Education, Upper Saddle River, New Jersey 2007.
[3] S. J. Ko and Y. H. Lee,“Center weighted median filters and their applications to image enhancement,”IEEE Transactions on Circuits and Systems, vol. 38, pp. 984-933, 1991.
[4] H. K. Kwan,“Fuzzy filters for noisy image filtering,”ISCAS 03.Proceedings of the 2003 International Symposium on Circuits and System, vol. 4, pp. 161-164, 2003.
[5] T. C. , K. K. ,and L. H. Chen “Tri-State median filter for image denoising,”IEEE Transaction on Image Processing, vol. 8, No. 12. pp. 1834-1838, 1999.
[6] E. Abreu and S. K. Mitra,“A signal-dependent rank ordered mean(SD-ROM)filter,”Proceedings of IEEE International Conference on Acoustics, Speech and Signal Processing, ICASSP-95, Detroit, pp. 2371-2374, 1995.
[7] T. Chen and H. R. Wu,“Adaptive impulse detection using center-weight median filter,”IEEE Signal Proceeding Letters, vol. 8, no1, Jan .2001.
[8] D. Zhang and Z. Wang,“Impulse noise detection and removal using fuzzy techniques,”Electrom.Letters, vol. 33, no. 5, pp. 378-319, 1997.
[9] Kenny K. and V. T, ,“Cluster-Based Adaptive Fuzzy Switching Median Filter for Universal Impulse Noise Reduction,” IEEE Transactions on Consumer Electronics,Vol. 56, No. 4, pp .2560-2568, Nov. 2010.
[10] Y. Z. , Q. H. T. and W. D. J. ,“Adaptive Fuzzy Median Filter for Images Corrupted by Impulse Noise,”IEEE Transactions on Consumer Electronices, vol. 54, no. 4, pp. 1920-1927, Nov. 2008.
[11] P. Civicioglu,“Removal of random-valued impulse noise from corrupted images,”IEEE Transactions on Consumer Electronics, Vol. 55, pp. 2097-2104, Nov.2010.
[12] F. Russo and G. Ramponi,“A fuzzy filter for image corrupted by impulse noise,”IEEE Signal Process.Lett,.vol. 3, no. 6, pp. 168-170, June 2006
[13] P. E. Ng and K. K. Ma,“A Switching median filter with boundary discriminative noise detection for extremely corrupted image, ”IEEE Transactions .Image Process., vol. 15, no. 6, pp. 1506-1516, June 2006.
[14] Y. Dong and S. Xu,“A new directional weighted median filter for removal of random-valued impulse noise,”IEEE Signal Process. Lett., vol. 14, no.3,
pp. 193-196, Mar.2007.
[15] J. Zhang,“An efficient median filter based method for removing radom-valued impulse noise,”Digit,signal Process., vol. 61, pp. 551-555, 2007.
[16] A. Fabijanska and D. Sankowski, “Noise adaptive switching median-based filter for impulse noise removal from extremely corrupted images”, IET image processing, pp. 472-480 July 2010.
[17] http://homepages.cae.wisc.edu/~ece533/images/
[18] 劉震昌，「數位影像處理」，台北，高立圖書有限公司，2010.
[19] 謝政勳、廖珗洲、李聯旺，「人工智慧：智慧型系統導論 第三版」台北，全華圖書有限公司，2012.