JPEG XR 是一種影像壓縮演算法和文件格式，由微軟在2006 年提出，其編碼過程中會使用到兩種轉換運算，分別為POT(Photo Overlap Transform) 和PCT(Photo Core Transform)。我們使用微軟提供的DPK(Device Porting Kit) 對圖片進行JPEG XR 編碼，發現在量化參數(Quantization Parameter) 設定為>1，也就是lossy 編碼的條件下，轉換後的圖片會產生棋盤現象(Checkerboard block artifacts)、邊框現象(Border artifacts) 及缺角現象(Corner artifacts)。在本論文中，我們針對上述這三個現象的成因進行分析，發現主要的問題在於POT 轉換和量化的步驟，為了解決此問題，我們提出一個“均勻矩陣(Normalization matrix)”的方法，在POT(encode) 步驟之後，和inverse-POT(decode) 步驟之前，額外乘上我們設計的均勻矩陣，藉由此方法，可簡單有效的改善棋盤、邊框及缺角現象，提供JPEG XR 在lossy 編碼下更佳的影像品質。

JPEG XR is a still-image compression standard and file format, proposed by Microsoft in 2006. Its encoding process utilizes two types of transform operations: POT (Photo Overlap Transform) and PCT (Photo Core Transform). Using the DPK (Device Porting Kit) provided by Microsoft, we performed encoding and decoding process of JPEG XR on images, and discovered that when the quantization parameter is >1, aka lossy compression conditions, checkerboard block artifacts, border artifacts and corner artifacts can be found in the resulting image. In this paper, we will analyze the cause of the above artifacts. It was found that the main problem lied in the step of POT and quantization. To solve this problem, we will propose the use of a “Normalization Matrix”. After POT (encoding) and before inverse-POT (decoding), we will add an extra step to multiply this normalization matrix. It is an easy and effective way to decrease checkerboard, border, and corner artifacts, thereby improving the image quality of lossy encoding JPEG XR.

[1] S. Srinivasan, C. Tu, S. L. Regunathan, and G. J. Sullivan, “HD Photo: A new image coding technology for digital photography,” in Proc. SPIE Appl. of Digital Image Processing XXX, vol. 6696, (San Diego, CA USA), p. 66960A, Aug. 2007.
[2] C. Tu, S. Srinivasan, G. J. Sullivan, S. Regunathan, and H. S. Malvar, “Lowcomplexity hierarchical lapped transform for lossy-to-lossless image coding in JPEG XR/HD Photo,” in Proc. of SPIE, vol. 7073, p. 70730C, 2008.
[3] W. Sweldens, “The Lifting Scheme: A new philosophy in biorthogonal wavelet constructions,” in Proc. SPIE Wavelet Applications in Signal and Image Processing III, vol. 2569, (San Diego, CA USA), pp. 68–79, July 1995.
[4] M. Corp., “HD Photo Device Porting Kit 1.0.” http://www.microsoft.com/en-us/download/details.aspx?id=5863, Nov. 2006.
[5] W. Sweldens, “The Lifting Scheme: A construction of second generation wavelets,” SIAM Journal on Mathematical Analysis, vol. 29, no. 2, pp. 511–546, 1997.
[6] Z. Wang, A. C. Bovik, H. R. Sheikh, and E. P. Simoncelli, “Image quality assessment: From error visibility to structural similarity,” Image Processing, IEEE Transactions on, vol. 13, no. 4, pp. 600–612, 2004.
[7] Z. Wang, A. C. Bovik, H. R. Sheikh, and E. P. Simoncelli, “The SSIM Index for Image Quality Assessment.” https://ece.uwaterloo.ca/~z70wang/research/ssim/, 2009.
[8] A. of Motion Picture Arts and S. (”A.M.P.A.S”), “The 84th academy awards poster.” http://www.oscars.org/academy/posters-books/posters/oscar.html, 2011.
[9] J. Gallaway, “james joyce(sketch from a famous photo of james joyce).” http://www.flickr.com/photos/63923085@N00/2304643567, 1998. owner:http://www.flickr.com/photos/osmium/.
[10] W. Dai, L. Liu, and T. D. Tran, “Adaptive block-based image coding with pre-/post-filtering,” in IEEE Data Compression Conference, (Snowbird, UT),pp. 73–82, Mar. 2005.
[11] F. De Simone, M. Ouaret, F. Dufaux, A. G. Tescher, and T. Ebrahimi, “A comparative study of JPEG2000, AVC/H.264, and HD photo,” in Proc. SPIE Appl. of Digital Image Processing XXX, vol. 6696, (San Diego, CA USA), 2007.
[12] F. Dufaux, G. Sullivan, and T. Ebrahimi, “The JPEG XR image coding standard [Standards in a Nutshell],” Signal Processing Magazine, IEEE, vol. 26, no. 6, pp. 195–199, 2009.
[13] Y. Gao, D. Chan, and J. Liang, “JPEG XR optimization with graph-based soft decision quantization,” in IEEE International Conference on Image Processing,(Brussels), pp. 313–316, Sep. 2011.
[14] H. Malvar, “Biorthogonal and nonuniform lapped transforms for transform coding with reduced blocking and ringing artifacts,” Signal Processing, IEEE Transactions on, vol. 46, no. 4, pp. 1043–1053, 1998.
[15] S. Srinivasan, C. Tu, Z. Zhou, D. Ray, S. L. Regunathan, and G. J. Sullivan, “An introduction to the HD Photo technical design,” JPEG document WG1 N4183, 2007.
[16] T. Tran, J. Liang, and C. Tu, “Lapped transform via time-domain pre-and postfiltering,” Signal Processing, IEEE Transactions on, vol. 51, no. 6, pp. 1557–1571, 2003.
[17] T. Tran, L. Liu, and P. Topiwala, “Performance comparison of leading image codecs: H. 264/AVC Intra, JPEG2000, and Microsoft HD Photo,” in Proc. SPIE, vol. 6696, p. 66960B, 2007.