臺灣博碩士論文加值系統

JPEG影像格式在一般生活上，例如網際網路或數位相機，是非常常見的。JPEG是一種影像壓縮格式標準，藉由壓縮可以縮減資料量的大小，以方便我們儲存或傳輸，但壓縮後會產生惱人的方塊效應，一塊塊的方塊雜訊讓視覺觀感上會非常的不舒服，這些因為壓縮所造成的方塊雜訊是不可避免的，方塊雜訊的存在又會影響到使用者的視覺感受，因此方塊雜訊的移除是必須且重要的。本文著重在JPEG影像方塊雜訊的去除，這裡所討論的雜訊是影像的離散餘弦轉換係數經由量化造成資料遺失所引起的方塊雜訊。傳統的濾波器方法，往往是先判斷所在的工作區塊是屬於平坦區域或是邊緣區域，然後再各自用不同的濾波器下去做處理。不同以往的濾波器設計，我們提出兩種處理雜訊的方法，第一種濾波器是基於模糊規則(fuzzy rule-based)的濾波器，此濾波器是一種權重平均的輸出，設計是基於三個影響系統輸出的參數:像素間灰階值的差距，像素間的距離、方向，以及處理區域之像素間的變異性。經由最小的平均誤差平方(LMS)演算法，我們可以得到此濾波的歸屬(權重)函數; 第二種我們則只著重處理方塊邊界部份的平滑處理，對於方塊效應不明顯的像素我們不做任何改變。最後由實驗結果我們可以證明我們所提出的方法在方塊雜訊移除的效果及其強健性。

The image compression by JPEG is very common in our daily life, for example, Internet or the digital camera. JPEG, itself, is a standard of image compression to decrease the data rate. Unfortunately, annoying blocking artifacts would appear. The blocking noise makes us uncomfortable usually. It is necessary and important to remove the blocking noises. The thesis introduces several schemes for the JPEG blocking noise removal: the noise discussed here is the blocking noise resulted from the quantization of DCT coefficients. Traditionally, different filters are applied respectively to the monotone area and the edge area, aiming at smoothing the monotone area or enhancing the edge area. We propose two methods to remove the blocking noise. The first one is the use of the fuzzy rule-based filter (FRB). The fuzzy rule-based filter’s output is a weighted average the processing pixel itself and its neighborhood pixels, dependent on the gray level difference between pixels, the spatial distance and direction between pixels, and the variance in the local window. Using the LMS learning algorithm we can determine the best membership function for the FRB filter. For the second method, we only deal with pixels around the block boundaries, letting the pixels not so blocky. By the simulation results, we have demonstrated the effectiveness and robustness in blocking noise removal our proposal schemes.

ABSTRACT (CHINESE) i
ABSTRACT (ENGLISH) ii
ACKNOWLEGEMENTS iv
CONTENTS v
LIST OF FIGURES vii
LIST OF TABLES x
CHAPTER 1 Inroduction 1
1.1 Overview 2
1.2 Thesis Outline 3
CHAPTER 2 Blocking Effect 4
2.1 JPEG Standard 4
2.2 Blocking Noise Types 10
CHAPTER 3 Algorithms 12
3.1 Nonlinear Space-Variant Filter 12
3.2 Blocking Effect Reduction of JPEG Images by Signal
Aptive Filtering 14
3.3 Adaptive Postfiltering of Transform Coefficients 18
3.4 Fuzzy Rule-Based Filters 21
3.4.1 Principle of Fuzzy Rule-Based Filters 22
3.4.2 Designing of Fuzzy Rule-Based Filters 26
3.4.3 Design of the Simplified FRB Filter 31
3.5 Boundary Extension 34
CHAPTER 4 Simulation Results 37
4.1 SNR 37
4.2 A Generalized Block-Edge Impairment Metric 38
4.3 Simulation Results 40
CHAPTER 5 Conclusion 64
References 65

[1] B. Ramamurthi and A. Gersho, “Nonlinear space variant postprocessing of block coded images,” IEEE Trans. Acoust., Speech, Singal Processing, vol. ASSP-34, pp. 1258-1267, 1986.
[2] Y. L. Lee, H. C. Kim, and H. W. Park, “Blocking effect reduction of JPEG images by signal adaptive filtering,” IEEE Trans. Image Processing, vol. 7, pp. 229-234, February 1998.
[3] Y. Yang, N. Galatsanos, and A. Katsaggelos, “Projection-based spatially adaptive reconstruction of block-transform compressed images,” IEEE Trans. Image Processing, vol. 4, pp. 896-908, July 1995.
[4] W. B. Pennebaker and J. L. Mitchell, JPEG Still Image Data Compression Standard. NEW YORK: Van Nostrand Reinhold, 1993.
[5] M. A. Sid-Ahmed, Image Processing, New York: McGraw-Hill, 1994.
[6] Programs for Digital Signal Processing. New York: IEEE Press, 1979.
[7] W. K. Pratt, Digital Image Processing, New York: Wiley, 1991.
[8] X. You and G. Crebbin, “A robust adaptive estimator for filtering noise in images,” IEEE Trans. Image Processing, vol. 4, pp. 693-699, May 1995.
[9] T. Chen, H. R. Wu, and B. Qiu, “Adaptive postfiltering of transform coefficients for the reduction of blocking artifacts” IEEE Trans. Circuits Syst. Video Technol., vol. 11, pp. 594-601, May 2001.
[10] K. Arakawa, “Fuzzy rule-based signal processing and its application to image restoration,” IEEE Journal on Selected Areas in Communications, Vol. 12, no. 9, Dec. 1994, pp. 1495-1502.
[11] H. L. Van Trees, Detection Estimation, and Modulation Theory. New York: Wiley, 1968.
[12] B. Widrow et al., “Adaptive noise canceling: Principles and applications,” Proc. IEEE, vol. 63, no. 12, pp. 1692-1716, Dec. 1975.
[13] H. R. Wu and M. Yuen, “A generalized block-edge impairment metric for video coding,” IEEE Signal Processing Letters, vol. 4, pp. 317-320, Nov. 1997.
[14] Y. Yang, N. P. Galatsanos, and A. K. Katsaggelos, “Regularized reconstruction to reduce blocking artifacts of block discrete cosine transform compressed images,” IEEE Trans. Image Processing, vol. 3, pp. 421-432, Dec. 1993.