科技的進步帶動了多媒體的發展，傳統的文字檢索技術已無法滿足使用者的需求，龐大的多媒體資料正以驚人的速度被製造出來。每天都有許多人在網路上找尋和瀏覽多媒體資料，為了能提供一個有效率的多媒體檢索技術，除了利用文字檢索以外，採用以內容為基礎(content based)之影像檢索系統被認為是目前對影像檢索最有效的方式，其方法是透過分析與影像，如形狀(shape)、顏色(color)和紋理(texture)等之低階特徵值，以便有效的檢索出相似的影像。
　　由於小波轉換(wavelet transform)具有良好的空間-頻率域的局部特性，因此在分析紋理方面非常成功，不但可以減少特徵值數目，而且具有代表性。小波轉換亦具有多重解析度的特質使得各個次頻道間互相獨立，相當有利於利用特徵值來作影像之比對。此外顏色紋理有強調顏色在空間上分佈的特性，而採用小波轉換所取得的顏色紋理特徵，可用以解決一般僅用顏色特徵值作影像之比對，而特徵向量過大的問題。
　　本論文考慮小波轉換後頻道間的相關性，將HL、LH二個次頻道視為單一特徵值，不但可以減少特徵向量，而且所取特徵值在搜尋上具有抗旋轉能力。本論文利用小波係數之特徵作為資料庫之索引，以輸入影像為中心來搜尋相似影像，能有效的縮短檢索回應時間。由採用小波係數之顏色紋理來強調影像資料庫顏色分佈之群聚性，實驗證明此種結合顏色紋理特徵之索引比對法可以增進影像搜尋系統之效能。

　　With the development of the World Wide Web(WWW) and fast computer technology, the use of visual information has become routine in scientific and commercial applications. Every day, large numbers of people are using the Internet for searching and browsing through diverse multimedia database.Due to the limitation of textual annotation based search , the content-based retrieval has been given more attention in recent years.It provides methods to query image database using image features as the basis for the queries. These image features include color, texture and shape of objects and regions.
　　Wavelet transform, because of its space-frequency localization characteristics, is preferred in many image and audio processing applications. Wavelet transform provide good multiresolution analytical tools for texture classification and it can achieve a high accuracy rate.
　　Considering the subband relationship after Wavelet transform, the essay suggests that we can combine HL and LH to a single feature which can decrease the vector numbers and the feature value has the rotation-invariant capability. By take the advantage of database indexing search ability ,the similar images using the input image condition and it reduces the querying response time.Besides, combining color texture features to emphasize the color distribution of image make retrieval system proves that it really can improve more efficiency.

參考文獻
[1] C. Faloutsos,, “Signature Based Text Retrieval Methods: A Survery, “ IEEE Data Eng. Bull.,vol. 13, no. 1, pp. 25-32, Mar.1990.

[2] Greg Pass, Ramin Zabin, and Justin Miller, “Comparing images using color coherence vectors”, In Proceedings of ACM Multimedia 96, pages 65—73, Boston MA USA, 1996.

[3] Markus Stricker and Markus Orengo. Similarity of color images. In Proc. SPIE Storage and Retrieval for Image and Video Databases, 1995.

[4] Gonzalez, R. M. and R. E. Woods. 1992. Digital image processing, 506-514. U. S. A,. : Addison-Wesley.

[5] Robert M. Haralick, K. Shanmugam, and Its’hak Dinstein. “Texture feature for image classification.” IEEE Trans. On Sys, Man, and Cyb, SMC-3(6), 1973.

[6] Haralick, R. M., K. Shanmugam and I. Dinstein. 1973. “Textural features for image classification .” IEEE Trans. SMC: 610-621.

[7] M. B. Henke-Reed and S. N. C. Cheng, “Cloth texture classification using the wavelet transform,” Journal of Imageing Science and Technology, vol 27, pp. 610, Dec. 1993.

[8] T. Chang and C. C. J. Kuo, “Texture analysis and classification with tree-structured wavelet transform, “ IEEE Trans. Image Processing, vol. 2, pp 429-441, Oct. 1993.

[9] X. Tang and W. K. Stewart, “Texture classification using principle-component analysis texhniques,” Proc. Image and Signal Processing for Remote Sensing, Rome, Italy, 1994.

[10] S. C. Tan and J. Kittler. “On colour texture representation and classification. “In Proc. Of the 2nd Int. Conf. on Image Processing, pages 390-395, 1992.

[11] D.K. Panjwani and G. Healey.” Markov random field models for unsupervised segmentation of textured color images. “IEEE Trans. Pattern Anal. Machine Intell., 17
(10): 939-954, 1995.

[12] Albuz, E.; Kocalar, E.; Khokhar, A.A “scalable color image indexing and retrieval using vector wavelets” Knowledge and Data Engineering, IEEE Transactions on , Volume: 13 Issue: 5 , Sept.-Oct. 2001 Page(s): 851 -861

[13] W. Niblack, R. Barber, and et al. “ The QBIC project: Querying images by content using color, texture and shape”, In Proc. SPIE Storage and Retrieval for Image and Video Databases, Feb 1994.

[14] Gupta, "Visual Information Retrieval: A Virage Perspective," 1995

[15] J. R. Smith and S. F. Chang,” VisualSEEk: A fully automated content-based image query system, “Proc. Fourth ACM International Multimedia Conference, Boston, MA, Nov. 1996.

[16] G. Wyszecki and W. S. Stiles.” Color Science, Concepts and Methods,” Quantitative Data and Formulas, 2nd ed. J. Wiley and Sons, New York, 1982.

[17] Pun, C.-M.; Lee, M.-C. “Rotation-invariant texture classification using a two-stage wavelet packet feature approach “ Vision, Image and Signal Processing, IEE Proceedings- , Volume: 148 Issue: 6 , Dec. 2001 Page(s):
422 -428

[18] Xiaoou Tang and W.Kenneth Stewart“Texture classification Using Wavelet packet and Fourier Transforms”

[19] R. Porter and N. Canagarajah Robust rotation-invariant texture classification: Wavelet, Gabor filter and GMRF based schemes” Vision Image Signal Process., IEE Proceedings-. Volume: 144 No. 3, June1997