臺灣博碩士論文加值系統

影像在擷取與傳輸的過程中，常常會因為外界環境與器材內部元件的影響而有非自然的資訊產生，這些資訊我們皆稱為影像雜訊，其中又以脈衝雜訊污染影像的程度最大。影像被雜訊污染會影響許多的後續處理，例如影像切割，人臉辨識，影像壓縮等，因此，將影像中的雜訊衰減，進而復原影像的自然資訊是一件刻不容緩的要務。基於一個5×5的工作遮罩，我們統計中心點像素與其他鄰居像素的像素關聯性，再根據像素關聯性的變化，推導出適應性鑽石型遮罩與適應性臨界值；另外，我們利用中心點像素與鄰近像素的同質性與相似性，研發出一套有效的脈衝雜訊偵測演算法，並根據像素關聯性，對脈衝雜訊演算法給予適當的加權值，再搭配適應性臨界值，進而判斷中心點像素是否為影像雜訊，如果為雜訊，再使用適應性鑽石型遮罩做中間值濾波器，取得最接近原始自然影像的像素值，取代被雜訊污染的像素值。我們的脈衝雜訊衰減濾波器由於是根據最大像素關聯性而來，因此，我們稱之為適應性最大像素關聯性中間值濾波器；由實驗模擬結果得知適應性最大像素關聯性中間值濾波器比許多已知的濾波器衰減影像脈衝雜訊的效果更好。

Images are often corrupted by impulse noise due to a noisy sensor or channel transmission errors. Impulse noise seriously affects the performance of image processing techniques, e.g. pattern recognition, image segmentation and image compression. To restore a high quality of image from a corrupted one with impulse noise is most important for image processing. Based on a square n×n working window, the statistics of the correlations between the center pixel and its neighbor pixels is computed to invent adaptive diamond window and adaptive threshold in an image. We infer an effective weighted decision-based median filtering algorithm with an adaptive threshold for noise detection from the statistics of the pixel-correlations based on an adaptive diamond window. If the pixel is corrupted by impulse noise, the median filter with an adaptive diamond window is used to reduce noise for image restoration. The decision-based median filtering algorithm to reduce noise and preserve noise-free pixel for image restoration is called adaptive highly pixel-correlation median filter as a result of the statistics of the correlations between the center pixel and its neighbor pixels based on a n×n working window in an image. The experimental results show that adaptive highly pixel-correlation median filter is more robust than many other well-known decision-based median filtering algorithms.

中文摘要 i
ABSTRACT ii
Acknowledgement iii
Contents iv
Lists of Tables vi
Lists of Figures vii
Chapter 1 Introduction 1
1.1 Background 1
1.2 Motivation 1
1.3 Organization of This Thesis 4
Chapter 2 Introduction to Impulse Noise 5
2.1 Fixed-valued Impulsive Noise 5
2.2 Random-valued Impulsive Noise 6
Chapter 3 Filters for Reducing Image Impulse Noise 8
3.1 Differential Rank Impulse Detector 9
3.2 Adaptive Two-Pass Median Filtering 10
3.3 Conditional Signal-Adaptive Median Filter 11
3.4 MSM Filter 13
Chapter 4 The Proposed Method 16
4.1 Pixel-Correlation 17
4.2 Diamond Working Window 21
4.3 Image Information Estimation 25
4.4 Adaptive Threshold 34
4.5 Impulse Noise Detection and Reduction 35
4.5.1 The Weighted Function 35
4.5.2 Impulse Noise Detection and Reduction 36
Chapter 5 The Experimental Results 39
Chapter 6 Conclusions 46
Reference 47
Publication List 50

[1] Shi Guangmin and Li Fudong, “ Image Denoising with Optimized Subband Threshold ”, Proceedings of the Fifth International Conference on Computational Intelligence and Multimedia Applications (ICCIMA’03), pp. 160-164, Sept. 2003.
[2] G. R. Arce and M. P. McLoughlin, “Theoretical analysis of the MAX/Median filter,” IEEE Trans. Acoust., Speech, Signal Processing, vol. ASSP-35, pp. 60-69, Jan. 1987.
[3] M. Gabbouj, E. J. Coyle, J. Neal, and C. Callagher, “An overview of median and stack filtering,” Circuits Syst. Signal Process., vol. 11, pp. 7-45, 1992.
[4] Y. H. Lee and S. Tantaratana, “Decision-based order statistics filters,” IEEE Trans. Acoust., Speech, Signal Processing, vol. 38, pp. 406-420, Mar. 1990.
[5] Piotr S. Windyga, “Fast Impulsive Noise Removal,” IEEE Transactions on Image Processing., vol. 10, pp. 173-179, Jan. 2001.
[6] G. Pok, Jyh-Charn Liu, and A. S., ”Selective removal of impulse noise based on homogeneity level information,” IEEE Trans. Image Processing, vol. 12, pp. 85-92, Jan. 2003.
[7] I. Aizenberg and C. Butakoff, “ Effective Impulse Detector Based on Rank-Order Criteria ”, IEEE Signal Processing Lett., Vol. 11, pp. 363 - 366, Mar. 2004.
[8] X. Xu, E. L. Miller, Dongbin Chen, and M. Sarhadi, “ Adaptive two-pass rank order filter to remove impulse noise in highly corrupted images ”, IEEE Trans. Image Processing, Vol. 13, pp. 238 - 247, Feb. 2004.
[9] Tao Chen and Hong Ren Wu, “ Space Variant Median Filters for the Restoration of Impulse Noise Corrupted Images ”, IEEE Transactions on Circuits and Systems II: Analog and Digital Signal Processing, Vol. 48, Issue 8, pp. 784 - 789, Aug. 2001.
[10] S. J. Ko and Y. H. Lee, “Center weighted median filters and their applications to image enhancement,” IEEE Trans. Circuits Syst., vol. 38, pp. 984-993, 1991.
[11] Roman Garnett, Timothy Huegerich, Charles Chui, and Wenjie He, “ A Universal Noise Removal Algorithm With an Impulse Detector ”, IEEE Transactions on Image Processing, Vol. 14, Issue 11, pp. 1747-1754, Nov. 2005.
[12] Rafael C. Gonzalez and Richard E. Woods, “ Digital Image Processing “, Second Edition, Prentice Hall, 2002.
[13] Rastislav Lukac, Bogdan Smolka, Karl Martin, Konstantinos N. Plataniotis and Anastasios N. Venetsanopoulos, “ Vector Filtering for Color Imaging ”, IEEE Signal Processing Magazine, Vol. 22, Issue 1, pp. 74-86, Jan. 2005.
[14] How-Lung Eng and Kai-Kuang Ma, “ Noise adaptive soft-switching median filter ”, IEEE Transactions on Image Processing, Vol. 10, Issue 2, pp. 242 - 251, Feb. 2001.
[15] Raymond H. Chan, Chung-Wa Ho and Mila Nikolova, “ Salt-and-Pepper Noise Removal by Median-Type Noise Detectors and Detail-Preserving Reqularization”, IEEE Transactions on Image Processing, Vol. 14, Issue 10, pp. 1479-1485, Oct. 2005.
[16] Lin Yin, Ruikang Yang, Moncef Gabbouj and Y. Neuvo, “ Weighted median filters: A tutorial ”, IEEE Transactions on Circuits and Systems, Vol. 43, Issue 3, pp. 157-192, March 1996.
[17] Shawn Jackson and Andreas Savakis, “ Adaptive Multilevel Median Filtering of Image Sequences ”, International Conference on Image Processing (ICIP 2005), Vol. 3, pp. III-545-8, Sept. 2005.