為了減少儲存的資訊量和網路傳輸的負載，影像壓縮已被廣泛的應用在多媒體通訊上。然而，壓縮會在影像的邊緣附近產生震盪型式的雜訊。我們將這個現象稱之為環效應。許多去除環效應的演算法被提出但無法同時去除環效應並且保留影像的細節。本論文提出一個嶄新的去環效應演算法。我們的演算法在濾波前對環效應進行偵測以保留影像細節。根據區塊中的像素以及其鄰近的區域，具有環效應的區塊被偵測出來。幅度濾波器被應用於具有環效應的區塊以去除環效應。實驗的結果說明了我們的演算法可以正確的偵測出環效應的位置而去環效應的結果在主觀的畫質和客觀的數據上來說比起其他的方法更好。

Data compression is commonly used in images in order to reduce the storage and transmission costs. However, compression introduces spurious oscillations known as ringing artifacts near major edges. Many algorithms were proposed to solve this problem but fail to preserve image details while removing ringing artifacts. In this thesis, we propose a novel adaptive algorithm to reduce ringing artifact in compressed images. In order to preserve image details, the proposed algorithm first detects the position of ringing artifact before filtering. The blocks with ringing artifact is detected according to each pixel in the block and its neighboring regions. Range filter is applied to the detected blocks for image deringing. Simulations results show that the proposed deringing algorithm can correctly detect the position of ringing artifact and effectively reduce the ringing artifacts while preserving image details and achieve better subjective and objective image quality than other recent approaches.

[1] ISO/IEC 14496-2. Information technology v generic coding of audiovisual objects v part 2: Visual. 2nd Ed., Appendix F V Post V processing for coding noise reduction, 2001.
[2] B. D. Liu C. B. Wu and J.F. Yang. E cient dct-domain blind measurement and reduction of blocking artifact. IEEE Transactions Circuits System Video Technology, 2002.
[3] Manduchi C. Tomasi, R. Bilateral ltering for gray and color images. Proc. 6th International Conference Computer Vision, 1998.
[4] Weisi Lin Xiaokang Yang Wenjun Zhang Guangtao Zhai, Jianfei Cai. Image deringing using quadtree based block-shift ltering. IEEE International Symposium on Circuits and Systems, 2008.
[5] Sehoon Yea Anthony Vetro Gung T. Vo, Truong Q. Nguyen. Adaptive fuzzy filtering for artifact reduction in compressed images and videos. IEEE Transactions on image processing, 2009.
[6] A. Vetro H. Sun K.E. Barner H-S Kong, Y. Nie. Coding artifact reduction using edge map guided adaptive and fuzzy lter. IEEE International Conference on Multimedian, 2004.
[7] A. Vetro H. Sun K.E. Barner H-S Kong, Y. Nie. Fast adaptive fuzzy post-filtering for coding artifact removal in interlaced video. IEEE International Conference on Acoustic, Speech and Signal Processing, 2005.
[8] A.J. Jerri. The gibbs phenomenon in fourier analysis, splines and wavelet approximations. Kluwer academic Publisher, Dordrecht, 1998.
[9] S. Liu and A. C. Bovik. Adaptive postprocessors with dct-based block classifications. IEEE Transactions Circuits System Video Technology, 2003.
[10] T. Q. Nguyen S. Yang, Y. H. Hu and D. L. Tull. Maximum-likelihood parameter estimation for image ringing artifact removal. IEEE Transactions Circuits System Video Technology, 2001.