在現今的影像壓縮系統中，除了減少影像存放空間之外，其他額外的功能(如可伸展性等)亦顯得越來越重要。JPEG2000 Part 1使用最大平移法(MAXSHIFT)實作有興趣區域編碼，這樣做的一個缺點是要使用額外的記憶空間來建立有興趣區域的遮罩。本論文提出一種基於CCSDS壓縮演算法為核心的有興趣區域編碼方法，不需要使用額外的記憶體產生遮罩，只需透過修改部分CCSDS的建議流程，來完成有興趣區域編碼這個功能。首先，我們會把小波轉換後的系數以CCSDS標準的建議重新安排所有小波系數的位置；再來依使用者輸入把影像分成數個區域，其中一個為有興趣區域，其餘為非有興趣區域；然後再按照使用者輸入的整體壓縮比來計算出不同區域所獲得分配的位元率；再把不同的區域按照被分配到的位元率以CCSDS為標準壓縮並輸出串流位元；最後把不同區域所代表的串流位元合併。

In the modern compression system, besides the reduction of the storage volume, others extra functional components (ex. scalability) become more and more important. JPEG2000 part I implements the region of interest (ROI) by MAXSHIFT, which needs to generate an ROI mask with extra memory space. This thesis proposes an ROI coding algorithm based on CCSDS kernel compression system, without extra memory space for generating mask. It just modifies parts of the procedure of CCSDS Recommended Standard. To accomplish the ROI coding function, first of all, we follow CCSDS Recommended Standard’s suggestion and rearrange the positions of coefficients transformed by discrete wavelet transform. Second, by the specification of user, we split several regions from original structure. One of these regions is ROI and others are Non-ROI. Based on the overall compression ratio inputted by the user, we re-calculate the different bit rate for such regions and then output the relative bit stream by CCSDS Recommended Standard compression. Finally, we merge the bit streams which are represented different regions in proper order.

[1] Amir Said and William A. Pearlman, “A new, fast, and efficient image codec based on set partitioning in hierarchical tress,” IEEE Trans. Circuits and Systems for Video Technology, Vol. 6, p243-250, June, 1996.

[2] Athanassios Skodras, Charilaos Christopoulos, and Touradj Ebrahimi,
“The JPEG2000 still image compression standard,” IEEE Signal Processing Mag.,
pp. 36–58, Sept. 2001.

[3] CCSDS Recommended Standard (Blue Book): Image Data Compression (CCSDS 122.0-B-1), Issue 1, Nov., 2005

[4] Charles K. Chui, “An Introduction to Wavelets, ” Academic Press, SanDiego,
ISBN 0585470901, 1992

[5] Collins T. and Atkins P., “Error-tolerant SPIHT image compression,”
IET Vision, Image Signal Processing, IEE Proceedings, Volume 148, Issue 3, p182-186, 2001

[6] Ekram Khan and Mohammed Ghanbari., “Error-detection and correction of transmission errors in SPIHT coded images,” IEEE Image Processing. 2002. Proceedings. 2002 International Conference on, Volume 2, pII-689-II-692, 2002

[7] Hong Pan, Wan-Chi Siu and Ngai-Fong Law, “A fast and low memory image coding algorithm based on lifting wavelet transform and modified SPIHT,” ELSEVIER Signal Processing: Image Communication 23, 146-161, 2008

[8] Jerome M. Shapiro, “Embedded image coding using zerotrees of wavelet coefficients,”IEEE Transactions on Signal Processing, Vol 41, No. 12, p. 3445-3462, 1993.

[9] Joel Askelöf, Mathias Larsson Carlander,Charilaos Christopoulos,
“Region of interest coding in JPEG 2000,” Elsevier Signal Processing: Image Communication 17, page 105-111, 2002.

[10] JPEG 2000 Part I: Final Draft International Standard (ISO/IECFDIS15444-1), ISO/IECJTC1/SC29/WG1 N1855, Aug. 2000.

[11] Kkeun-hyeong Park, Chul Soo Lee and Hyun Wook Park, “Region-of-interest coding
based on Set Partitioning In Hierarchical Trees,” Circuits and Systems for Video
Technology, IEEE Transactions on, Volume 12, Issue 2, p106-113, 2002

[12] Shih-Hsien Yang and Tsung-Chao Cheng, “Error-resilient SPIHT image coding,” IET Electronics Letters, Volume36, Issue 3, p 208-210, 2000