為了達到有效影像檢索之目的，我們提出一自動化前置處理方法來去除不必要的影像背景資訊，來避免其影響檢索之結果。針對背景色彩變異較平緩之影像，所提出之方法能有效擷取影像前景物件，並去除不必要之背景資訊。所提出之方法共包含兩階段：第一、多階型態學影像分割，第二、二維網格精煉。首先在第一階段時，利用多階重建開放算子和多階重建封閉算子來對影像進行分割，如此可得到較粗糙之影像前景物件，接著在第二階段時，利用二維網格之方法，對在第一階段所得之影像做內插法，以便獲得正確之影像前景物件。接著我們從分割過的影像中擷取MPEG-7的特徵值，然後將其嵌入至JPEG2000之位元串中，而所提出之方法能整合MPEG-7和JPEG2000之優點，進而提供一有效之影像檢索平台，此檢索系統是基於JPEG2000影像之格式而發展，並將MPEG-7特徵值嵌入影像之中，經由提出之方法所得之影像前景物件作為ROI遮罩，此ROI資訊將不會增加JPEG2000檔案格式之大小，亦不會影響傳輸效率，此外，實驗結果也證明經由之法，檢索效能可大幅提升，並表現於精確率-回取率曲線上。

An image retrieval system that provides intelligent preprocessing and accurate retrieval results is proposed in this thesis. To obtain accurate feature descriptions, an automatic preprocessing method is proposed to eliminate unnecessary background information of the images. For most images with the smooth color variation backgrounds, the foreground object could be extracted with very high accuracy by the proposed method which comprises two processing stages: (a) the foreground segmentation using the multiscale morphology and; (b) refining the segmented object boundary by a 2D mesh. At the first stage, we segmented the image using multiscale opening and closing by reconstruction to obtain the coarse foreground object of the image. The coarse object is further refined by a 2D mesh to obtain the accurate foreground boundaries in the second stage. Experiments demonstrate that the foreground object of the image could be identified efficiently and correctly. Then we extracted MPEG-7 recognized features from the segmented image and embedded them in the JPEG2000 bitstreams. This manipulation would help to combine MPEG-7 descriptions and JPEG2000 ROI display to provide a unified platform for image retrieval. The image search engine can then be developed on this unified platform. The information of segmented foreground objects can be embedded in the JPEG2000 codestream without increasing the required bandwidth. Simulations show that the preprocessing helps much in improving the precision-recall perfor- mance, as compared to that without preprocessing.

[1] K. Jeffay and H. Zhang, Readings in Multimedia Computing and Networking, San Francisco: Morgan Kaufmann, CA, 2002.
[2] G. K. Wallace, “The JPEG still picture compression standard,” Communications of the ACM, vol. 34, no. 4, pp. 30-44, Apr. 1991.
[3] D. S. Taubman and M. W. Marcellin, JPEG2000: Image Compression Funda- mentals, Standards and Practice, Norwell, MA: Kluwer Academic Publishers, 2002.
[4] JPEG2000 Part I: Final Draft International Standard (ISO/IEC FDIS15444-1), ISO/IEC JTC1/SC29/WG1 N1855, Aug. 2000.
[5] JPEG2000 Requirements and Profiles, ISO/IEC JTC1/SC29/WG1 N1271, Mar. 1999.
[6] L. Cieplinski et al, “MPEG-7 visual part of experimentation model version 11.1,” ISO/IEC JTC/SC29/WG11 MPEG01/M7691, 2001.
[7] D. S. Taubman, “High performance scalable image compression with EBCOT,” IEEE Trans. on Image Processing, vol. 9, no. 7, pp. 1158-1170, Jul. 2000.
[8] B. S. Manjunath, P. Salembier, and T. Sikora, Introduction to MPEG-7: Multi- media Content Description Interface, New York: Wiley, 2002.
[9] Z. Li and M. S. Drew, Fundamentals of Multimedia, New Jersey: Prentice Hall, 2004.
[10] Overview of the MPEG-7 standard, ISO/IEC JTC1/SC29/WG11, Doc.N3752, La baule, France, Oct. 2000.
[11] T. Sikora, “The MPEG-7 visual standard for content description - An overview,” IEEE Trans. on Circuits and Systems for Video Technology, vol. 11, no. 6, pp. 696-702, Jun. 2001.
[12] H. D. Cheng, X. H. Jiang, Y. Sun, and J. Wang, “Color image segmentation: advances and prospects,” Pattern Recognition, vol. 34, pp. 2259-2281, 2001.
[13] L. Vincent, “Watersheds in digital spaces: an efficient algorithm based on imm- ersion simulations,” IEEE Trans. Pattern Anal. Machine Intell., vol. 13, no. 6, pp. 583-598, 1991.
[14] J. M. Gauch, “Image segmentatioon and analysis via multiscale gradient watershed hierarchies,” IEEE Trans. on Image Processing, vol. 8, no. 1, Jan. 1999.
[15] Y. Deng and B. S. Manjunath, “Unsupervised segmentation of color-texture regions in images and video,” IEEE Trans. on Pattern Analysis and Machine Intelligence, vol. 23, no. 8, pp. 800-810, 2001.
[16] C. Christopoulos, A. Skodras and T. Ebrahimi, “The JPEG2000 still image coding system: an overview,” IEEE Trans. on Consumer Electronics, vol. 46, no. 4, pp. 1103-1127, Nov. 2000.
[17] C. S. Burrus, R. A. Gopmath, and H. Guo, Introduction to Wavelets and Wavelet Transforms, New Jersey: Prentice Hall, 1998.
[18] M. Antonini, M. Barlaud, P. Mathieu and I. Daubechies: “Image coding using the wavelet transform,” IEEE Trans. Image Proc., pp. 205-220, Apr. 1992.
[19] A. R. Calderbank, I. Daubechies, W. Sweldens and B. L. Yeo, “Lossless image compression using integer to integer wavelet transforms,” Proc. IEEE Int. Conf. Image Processing, Santa Barbara, USA, Oct. 1997.
[20] JPEG2000 Part I Final Committee Draft Version 1.0, ISO/ISC JTC 1/SC 29/WG 1 (ITU-T SG8), Mar. 2000.
[21] Z. Wang and A. C. Bovik, “Bitplane-by-Bitplane Shift (BbBShift) - A suggestion for JPEG2000 region of interest image coding,” IEEE Signal Processing Letters, vol. 9, no. 5, May 2002.
[22] Z. Wang, S. Banerjee, B. L. Evans and A. C. Bovik, “Generalized Bitplane- by-Bitplane shift method for JPEG2000 ROI coding,” Proc. IEEE Int. Conf. on Image Processing, vol. 3, pp. 81-84, Sep. 2002.
[23] L. Liu and G. Fan, “A new method for JPEG2000 Region-of-Interest image coding: most significant bitplanes shift,” in Proc. of the 45th IEEE Int. Midwest Symposium on Circuits and Systems, Tulsa, OK, Aug. 2002.
[24] L. Liu and G. Fan, “A new JPEG2000 Region-of-Interest image coding method: partial significant bitplanes shift,” IEEE Signal Processing Letters, vol. 10, no. 2, pp. 35-39, Feb. 2003.
[25] A. P. Bradley and F. W. M. Stentiford, “JPEG 2000 and region of interest coding,” Digital Image Computing Techniques and Applications (DICTA), Melbourne, Australia, pp. 303-308, Jan. 2002.
[26] J. Serra, Image Analysis Using Mathematical Morphology, London, U.K.: Academic, 1982.
[27] S. Mukhopadhyay and B. Chanda, “Multiscale morphological segmentation of gray-scale image,” IEEE Trans. on Image Processing, vol. 12, no. 5, May 2003.
[28] J. Serra and L. Vincent, “An overview of morphological filtering,” IEEE Trans. Circuits, Systems and Signal Proc., vol. 11, no. 1, pp. 47-108, Jan. 1992.