隨著網路越來越發達，如何在網際網路世界中找出想要的影像，一直是個重要的問題。以影像或影片之內容為基礎的檢索研究，已被廣泛的研究，也都各有各的優缺點。在以往，使用關鍵字進行影像或影片的檢索時，使用者需先輸入關鍵字，然後利用關鍵字來做檢索工作，但此種方法面臨兩個問題，第一個問題是資料庫越來越龐大，若想利用關鍵字來檢索，首先須對資料做關鍵字的註解，此過程是非常花時間的；第二問題是對同一資料而言，不同的人對相同之資料會有不同的註解，此會導致檢索之正確性不一致。本文提出一個影像檢索的方法，對影像提取其全域與區域特徵(例如：色彩、勒讓德矩、影像特徵點個數等)並利用影像的特徵點個數做判斷，對不同複雜度之影像，分別利用不同的方法來加以執行檢索工作。在影片內容之檢索方面，將影片經分鏡處理，提取其關鍵影格，利用這些關鍵影格來代表一段影片，之後應用影像檢索的方法來做影片內容之檢索工作，經由實際影像與影片測試之結果顯示，當影像或影片內容物具有許多外觀較平滑之物件存在時，本文所提出之方法具有很不錯之成功率。本系統對一般電影影片與卡通影片資料庫測試分別可以達到87%與84%之檢索率。

Keywords are usually used to search images from the Internet. It is usually needed to make notations for the image during the image data base was created. The description of the notations depends on the person who charges the job. Different person may have different notations. In this research, it was tried to propose a unify method for image retrieval system.
The proposed method is also applied to the video system. The input image is normalized and then quantized the color channels to reduce its dynamic region for improving the matching successful rate via the CIEL*a*b* color space. Both global and local features are considered. For image retrieving, the number of image feature points is used as the criterion for selecting one of the two matching method, feature vector, which combined with global features and region features, matching or Speeded-Up Robust Features (SURF) matching. The feature vector method will give a much better results than that of SURF for these images with smoothed content.
The proposed method is also applied to search the frames from a video. Firstly, the video is divided into a number of shots. Then the key frames of each shot are determined and normalized. These key frames are then processed as images. From the results shown, the proposed method can improve the performance for those images with smoothing contents. Both the cartoon database and video database were tested, it is shown that the proposed method get 84% and 87% retrieval rate, respectively.

[1] Suresh Pabboju and A.Venu Gopal Reddy, “A novel approach for content-based image indexing and Retrieval system using global and region feature,” International Journal of computer science and network security, Vol.9, No.2, pp.119-130, 2009
[2] Chuen-Horng Lin, Rong-Tai Chen, Yung-Kuan Chan, “A smart content-based image retrieval system based on color and texture feature,” Image and Vision Computing, Vol. 27, Issue 6, pp. 658-665, 2009
[3] Brunelli, R. and Mich, O., “Histograms analysis for image retrieval.” Pattern Recognition, Vol. 34, Issue 8, pp. 1625-1637, 2001
[4] Sunil Lee and Chang D. Yoo, “Robust video fingerprinting for content-based video identification,” IEEE Transactions on circuits and systems for video technology, Vol. 18, No. 7,pp. 983–988, 2008
[5] Khalid M.Hosny, “Efficient computation of Legendre moments for gray level images,” international journal of image and graphics, vol. 7, No. 4, pp. 735-747, 2007
[6] Cecilia Di Ruberto and Andrea Morgera, “Moment-based techniques for image retrieval,” Database and Expert Systems Application, 2008.19th International Conference , Bilbao Spain. 2008
[7] G.A. Papakostas, E.G. Karakasis, D.E. Koulouriotis, “Accurate and speedy computation of image Legendre moments for computer vision applications, image and vision computing,” Image and Vision Computing ,Vol. 28, Issue 3, pp. 414-423, 2010
[8] D.Lowe, “object recognition from local scale-invariant feature,” Computer Vision, The Proceedings of the Seventh IEEE International Conference , Corfu Greece , 1999
[9] D.Lowe, “Distinctive image features from scale-invariant keypoints, cascade filtering approach,” International Journal of Computer Visio, Vol. 60, No.2, pp.91-110, 2004
[10]Herbert Bay, Andreas Ess, Tinne Tuytelaars, and Luc Van Gool, “Speeded-up robust features(SURF),” computer vision and image understanding, vol. 110, issue 3, pp. 346-359, 2008
[11]Chun-Rong Huang, Chu-Song Chen, Pau-Choo Chung,“Contrast context histogram-An efficient discriminating local descriptor for object recognition and image matching,” Pattern Recognition, Vol. 41, Issue 10, pp. 3071-3077 , 2008
[12]Kuan-Ta Chen, Jau-Yuan Chen, Chun-Rong Huang, and Chu-Song Chen, “Fighting phishing with discriminative keypoint features,” IEEE internet computing, vol. 13, No. 3, pp. 56-63, 2009
[13]H.J. Zhang, A. Kankanhalli and S.Smoliar, “Automatic partitioning of video,” Multimedia system , 2002
[14]M-K. Hu, “Visual pattern recognition by moment invariants,” Information Theory, IEEE Transactions on, Vol. 8, No. 2, pp. 179-187, 1962
[15]A.Chiang, S.Liao, Q.Lu and M.Pawlak, “Gegenbauer moment-based applications for Chinese character recognition,” Electrical and Computer Engineering conference, Canada, 2002
[16]Ali Broumandnia and Jamshid shanbehzadeh, “Fast Zernike wavelet moments for farsi character recognition,” Image and Vision Computing,Vol. 25, Issue 5, pp. 717-726 , 2007
[17]Mukundan R. and Ramakrishnan K.R., “Moment Functions in Image Analysis:
Theory and Applications,” World Scientific, Singapore, 1998
[18]M law Pawlak, “Image analysis by moments: reconstruction and
computational aspects,” Oficyna Wydawnicza PolitechnikiWrocLawskiej
WrocLaw, 2006.
[19] Milton Abramowitz and Irene A. Stegun, eds. “Handbook of mathematical functions with formulas, graphs, and mathematical tables,” Courier Dover Publications, 1965
[20]Matthew Brown and David Lowe, “Invariant features from interest point
groups,” International Journal of Computer Vision, vol. 60, No.2, pp.
91-110, 2004
[21] Jose Antonio Martin H, Matilde Santos, Javier de Lope, “Orthogonal
variant moments features in image analysis,” Information
Sciences,Vol. 180, Issue 6, pp. 846-860, 2010

[22]P.W. Huang, S.K. Dai,”Image retrieval by texture similarity,” Pattern recognition , Vol. 36, No.3, pp. 665-679, 2003
[23]N.Jhanwar, S.Chaudhuruib, G.Seetharamanc, B.Zavidovique, “content based image retrieval using motif co-occurrence matrix,” Image and Vision Computing, Vol. 22, Issue 14, pp. 1211-1220, 2004
[24]Naveen K.Nishchal, Arun K.Gupta, Vinod K.Rastogi, “Retrieval and classification of objects using generic Fourier,Legendre moment, and wavelet Zernike moment descriptors and recognition using joint transform correlator,” Optics & Laser Technology, Vol. 40, Issue 3,pp. 517-527, 2008.
[25]J.J.Koenderink, “The structure of images,” Biological cybernetics, Vol. 50, No. 5, pp.363-370, 1984.
[26]T.Lindeberg, “Scale-space theory:A basic tool for analyzing structures at different scales,”Journal of applied statistics, Vol. 21, No. 2, pp.224-270, 1994.
[27] http://www.fotosearch.com.tw