本論文提出一種改善JPEG解壓縮影像品質方法，為Chang et al. 所提方法之改進版本。在JPEG壓縮影像時，利用一個改進係數調整區塊(Modified Coefficient Adjustment Block)記錄各區塊之DCT係數在量化步驟因四捨五入所造成誤差的情形，當影像解壓縮反量化DCT係數時，再依照紀錄的情形來調整重建的DCT係數，進而轉換成解壓縮影像，由於各影像屬性不同(即量化步驟中被四捨五入之小數值分佈不同)，因此本論文提供兩種參數選擇方式來調整係數，及固定與自適性調整參數。
實驗結果顯示，固定與自適性參數調整皆比Chang et al. 所提的方法有明顯的改進。

In this paper we proposed a method to enhance the quality of the decompressed JPEG images, which is an improved version of the method proposed by Chang et.al.. A so-called Modified Coefficient Adjustment Block is used to record the round off situation while DCT coefficients in a block are quantized according to a given quantization block. In the decompression stage, the information recorded in the Modified Coefficient Adjustment Block is then referred to adjust the de-quantized DCT coefficients before the Inverse DCT transformation. We provide two choices of parameter setting, fixed and adaptive parameters. Experimental results show that no matter whether fixed or adaptive parameters are used, our method provides significantly better results than The method proposed by Chang et al..

目錄
第一章 緒論	1
1.1 研究動機與目的	1
1.2 論文架構	3
第二章 相關文獻回顧	4
第三章 相關基礎理論	7
3.1 JPEG壓縮相關背景	7
3.2 PSNR (Peak Signal-to-Noise Ratio)	9
3.3 Coefficient Adjustment	10
第四章 研究方法	13
4.1 Modified Coefficient Adjustment Block	14
4.2 固定分割點與調整參數 (Fixed Segmenting Points and Adjusting Parameters)	17
4.3 自適性分割點與調整參數 (Adaptive Segmenting Points and Adjusting Parameters)	19
第五章 實驗結果與討論	23
第六章 結論	27
參考文獻	28
附錄：英文論文	30

圖目錄
圖1 標準量化表	8
圖2 Z字型掃瞄矩陣的順序	8
圖3 JPEG壓縮流程示意圖	9
圖4 調整位元表(Adjustment Bit Map)之範例	12
圖5 係數調整區塊(Coefficient Adjustment Block)之範例	12
圖6 方法流程圖	14
圖7 Modified Coefficient Adjustment Block之範例	16
圖8 小數範圍區分示意圖	16
圖9 以Lena圖為例統計落在[0, 0.5)內之小數分布直方圖	20
圖10 以Lena圖為例統計落在[0.5, 1)內之小數分布直方圖	20
圖11 部分測試影像	23
圖12 區塊中36個低中頻係數	24

表目錄
表1 區間分割值與調整參數取法組合	22
表2 Lena以不同組合計算之自適性調整參數	24
表3 Sena以不同組合計算之自適性調整參數	25
表4 House以不同組合計算之自適性調整參數	25
表5 Baboon以不同組合計算之自適性調整參數	25
表6 不同組合之自適性調整參數取法PSNR比較	26
表7 本篇論文方法與其他方法之比較	26

[1] D. Yu, F. Sattar and B. Barkat, “Multiresolution Fragile Watermarking using Complex Chirp Signals for Content Authentication,” Pattern Recognition, May 2006, Vol. 39, Issue 5, pp. 935-952.
[2] J. L. Liu, D. C. Lou, M. C. Chang, and H. K. Tso, “A Robust Watermarking Scheme using Self-Reference Image,” Computer Standards & Interfaces, Jan. 2006, Vol. 28, Issue 3, pp. 356-367.
[3] J. Fridrich, M. Goljan, and R. Du, “Detecting LSB Steganography in Color, and Gray-scale Images,” IEEE Multimedia, 2001, Vol. 8, No. 4, pp. 22-28.
[4] C. C. Chang, and H. W. Tseng, “A Steganographic Method for Digital images using Side Match,” Pattern Recognition Letters, Sep. 2004, Vol. 25, No.12, pp. 1431–1437.
[5] D. C. Wu, and W. H. Tsai, “A Steganographic Method for Images by Pixel-Value Differencing,” Pattern Recognition Letters, Jun. 2003, Vol. 27, No. 9-10, pp. 1613–1626.
[6] Min-Yen Chiu, Yu-Sheng Liao, and Jiun-Jian Liaw, “Improved Steganographic Technique for the Image Quality of PVD,” International Conference on Advanced Information Technologies, 2010, pp. 98.
[7] Reza Jafari, Ziou Djemel, and Mammeri Abdelhamid, “Increasing Compression of JPEG images Using Steganography,” Robotic and Sensors Environments (ROSE) IEEE, 2011, pp. 226 - 230 
[8] Chin-Chen Chang, Yung-Chen Chou, and Jau-Ji Shen, “Improving Image Quality for JPEG Compression Knowledge-Based Intelligent Information and Engineering Systems,” Lecture Notes in Computer Science Vol. 3683, 2005, pp. 442-448.
[9] G. K. Wallace, “The JPEG still picture compression standard,” IEEE Transactions on Consumer Electronics, 1992, Vol. 38, No. 1.
[10] D. A. Huffman, “A method for the construction of minimum redundancy codes,” Proc. IRE, 1952, Vol. 40, pp. 1098-l 101.