臺灣博碩士論文加值系統

隨著近年來網際網路的蓬勃發展，現今網路已成為現代人重要的資訊來源之一，因此如何快速且正確的搜索出所需要的資訊，在近年來便成為一門重要的課題，而使許多研究紛紛提出各種圖片檢索技術。其中一種常用的圖片檢索技術為以圖像內容作檢索的一種方法，稱為CBIR (Content-Based Image Retrieval)，但其具有低階圖片特徵無法完全表達高階語意的缺點 (semantic gap)，許多研究中皆提到相關式回饋技術(Relevance Feedback)為改善的方法之一。RF為CBIR系統在得到檢索結果後，讓使用者對Query圖片和檢索結果進行比較，並提供回饋以使下一次檢索結果較佳的方法，而虛擬相關式回饋技術(Pseudo Relevance Feedback)被提出，以電腦取代使用者以避免誤判情形發生。
基於PRF的假設，本研究提出一個演算法可以改良Feedback結果中，回饋資訊的影響力沒有隨著重要性而提升的缺點，稱為Block-Based Pseudo Relevance Feedback (BBPRF)。此演算法的概念為將Feedback結果中，每k個結果畫分為同一區間，並將結果乘以該區間的權重值，以所得到的新的結果做下一個回合(iteration)的回饋；該權重值隨著第一次檢索結果的排行遞減，以達到「排名越前面的結果，相對重要性越大」的目的。
本研究以常見的NUS-WIDE和Caltech256兩種資料集進行實驗，使用RF中常見的演算法之一Rocchio演算法進行比較。實驗一為探討本研究所出的改良演算法檢索結果是否有比傳統演算法的結果來的好，實驗二為加入使用者回饋進行檢索結果的探討。
實驗結果中，NUS-WIDE和Caltech256兩種資料集中，改良的演算法所得到的效果皆比傳統的虛擬相關式回饋演算法好，且過程所費時間並無過大差距。以此可看出本研究所提出之改良式的虛擬相關式回饋演算法具有更好的效能。

Nowadays the network has become one of the important ways to obtain information. Therefore, it is important to effectively search for information. For image search, CBIR (Content-Based Image Retrieval) is major technique. However, the semantic gap problem limits the performance of CBIR systems. In literature, RF (Relevance Feedback) can be used to improve the retrieval performance of CBIR systems. It is usually based on asking users to give feedbacks, and the retrieval results are re-ranked. One major limitation of RF is the need of the user in the loop process. To this end, PRF (Pseudo Relevance Feedback) was proposed that considers top-k images as the pseudo feedbacks to re-rank the retrieval results.
This thesis proposes an algorithm called Block-Based Pseudo Relevance Feedback (BBPRF) to improve the traditional PRF approach. The idea of this algorithm is to assign higher weights to higher ranked images. Particularly, top-k images as the feedbacks are divided into two to k blocks and each block has a specific weight, so the weighted feedbacks will benefit the next feedback iteration.
The experiments are based on the NUS-WIDE and Caltech256 datasets and the Rocchio algorithm is used as the traditional feedback algorithm. The first experimental results show that our proposed BBPRF performs better than the traditional PRF approach in terms of precision at 10, 20, and 50. In particularly, using top 30 images with 30 blocks perform the best. The second study further integrates the user’s feedbacks and BBPRF, and the retrieval performance is even better than using BBPRF alone.

第一章 緒論 1
1-1 研究背景 1
1-2 研究動機與目的 5
1-3 研究貢獻 7
1-4 論文架構 7
第二章 文獻探討 8
2-1 CBIR 8
2-2 使用者回饋相關技術 11
2-2.1 使用者回饋(Relevance Feedback) 11
2-2.2 Pseudo Relevance Feedback 12
2-2.3 Rocchio Algorithm 12
第三章 研究方法 14
3-1. Block-Based Pseudo Relevance Feedback (BBPRF) 14
3-2. 實驗步驟與流程 18
第四章 實驗分析 19
4-1 實驗設計 19
4-2 實驗一 20
4-2.1 NUS-WIDE 20
4-2.2 Caltech256 22
4-2.3 實驗一效能分析 24
4-3 實驗二 26
4-3.1 NUS-WIDE 26
4-3.2 Caltech256 28
4-3.3 實驗二效能分析 31
第五章 結論與未來研究方向 33
5-1 總結 33
5-2 未來研究方向 33
參考文獻 35
附錄 39

Androutsas, D. et al. (1998) Image retrieval using directional detail histograms. Storage and Retrieval for Image and Video Databases VI, Proc SPIE 3312, pp. 129-137.
Bian, W., and Tao, D. (2010) Biased Discriminant Euclidean Embedding for Content-Based Image Retrieval. IEEE Transactions on Image Processing, pp. 545-554.
Biederman, I. (1987) Recognition-by-components: a theory of human image understanding. Psychological Review, pp. 115-147.
Del Bimbo, A., Pala, P., and Santini, S. (1996) Image retrieval by elastic matching of shapes and image patterns. Proceedings of the Third IEEE International Conference on Multimedia Computing and Systems, pp. 215-218.
Carson, C., Belongie S., Greenspan H., and Malik J. (1997) Region-Based image querying. Proceedings of IEEE Workshop on Content-Based Access of Image and Video Libraries, pp. 42-49.
Chan, Y., and Kung, S.Y. (1997) A hierarchical algorithm for image retrieval by sketch. IEEE First Workshop on Multimedia Signal Processing, pp. 564-569.
Duan, L., Li, W., Tsang, I.W.H., and Xu, D. (2011) Improving web image search by bag-based reranking. IEEE Transactions on Image Processing, pp. 3280-3290.
Eakins, J., and Graham M. (1999) Content-based Image Retrieval. University of Northumbria at Newcastle.
Haralick, R.M., Shanmugam, K., and Dinstein, I. (1973) Texture features for image classification, IEEE Transactions on Systems, Man, and Cybernetics, pp. 610-621.
Hirata, K., and Kato, T. (1992) Query by Visual Example – Content-based Image Retrieval. Third International Conference on Extending Database Technology, pp. 56-71.
Jain, A.K., and Vailaya, A. (1996) Image retrieval using color and shape. Pattern recognition, pp. 1233-1244.
Kaplan, L.M., Murenzi, R., and Namuduri, K.R. (1998) Fast texture database retrieval using extended fractal features. Storage and Retrieval for Image and Video Databases VI, Proc SPIE 3312, pp. 162-173.
Kimia, B. et al. (1997) A shock-based approach for indexing of image databases using shape. Multimedia Storage and Archiving Systems II, Proc SPIE 3229, pp. 288-302.
Lee, K.S., Croft, W.B., and Allan, J. (2008) A cluster-based resampling method for pseudo-relevance feedback. ACM International Conference on Research and Development in Information Retrieval, pp. 235-242.
Liu, F., and Picard, R.W. (1996) Periodicity, directionality and randomness: Wold features for image modeling and retrieval. IEEE Transactions on Pattern Analysis and Machine Intelligence, pp. 722- 733.
Ma, W.Y., and Manjunath, B.S. (1998) A Texture thesaurus For Browsing Large Aerial Photographs. Journal of the American Society for Information Science, pp.633-648.
Manjunath, B.S., and Ma, W.Y. (1996) Texture features for browsing and retrieval of large image data. IEEE Transactions on Pattern Analysis and Machine Intelligence, pp. 837-842.
Manning, C.D., Raghavan, P., and Schütze, H. (2008) An Introduction to Information Retrieval. Cambridge University Press.
Mehrotra, R., and Gary, J.E. (1995) Similar-shape retrieval in shape data management. IEEE Computer, pp. 57-62.
Moschitti, A. (2003) A study on optimal parameter tuning for Rocchio text classifier. European Conference on Information Retrieval, pp. 420-435.
Niblack, W. et al. (1993) The QBIC project: querying images by color, texture and shape. IBM Research Report RJ-9203.
Pentland, A. et al. (1996) Photobook: tools for content-based manipulation of image databases. International Journal of Computer Vision, pp. 233-254.
Rui, Y., She, A.C., and Huang, T.S. (1996) Modified Fourier descriptors for shape representation—a practical approach, Proceedings of First International Workshop on Image Databases and Multi Media Search.
Rui, Y., Huang, T.S., and Chang, S.-F. (1999) Image Retrieval: Current Techniques, Promising Directions, and Open Issues. Journal of Visual Communication and Image Representation, pp. 30-62.
Smith, J.R., and Chang, S.F. (1995) Single color extraction and image query. International Conference on Image Processing, pp. 528-531.
Smith, J.R., and Chang, S.F. (1995) Tools and techniques for color image retrieval. IS & T/SPIE Proceedings.
Smith, J.R., and Chang, S.F. (1996) Automated binary texture feature sets for image retrieval. IEEE Transactions on Acoustics, Speech, and Signal Processing, pp. 2239-2242.
Stricker, M., and Orengo, M. (1995) Similarity of color images. Storage and Retrieval for Image and Video Databases III, Proc SPIE 2420, pp. 381-392.
Stricker, M., and Dimai, A. (1996) Color indexing with weak spatial constraints. Storage and Retrieval for Image and Video Databases IV, Proc SPIE 2670, pp. 29-40.
Swain, M.J., and Ballard, D.H. (1991) Color indexing. International Journal of Computer Vision, pp. 11-32.
Tamura, H., Mori, S., and Yamawaki, T. (1978) Texture features corresponding to visual perception, IEEE Transactions on Systems, Man, and Cybernetics, pp. 460-472.
Tirthapura, S. et al. (1998) indexing based on edit-distance matching of shape graphs. Multimedia Storage and Archiving Systems III, Proc SPIE 3527, pp. 25-36.
Varma, M., and Zisserman, A. (2005) A statistical approach to texture classification from single images. International Journal of Computer Vision, pp. 61-81.
Wang, Y.M., and Xu, Z.G. (2009) Image Retrieval Using the Color Approximation Histogram Based on Rough Set Theory. International Conference on Information Engineering and Computer Science, pp. 1-4.
http://bigsnarf.wordpress.com/2012/03/25/creating-my-first-algorithm-from-scratch-euclidean-distance-and-pearson-correlation/
http://lms.comp.nus.edu.sg/research/NUS-WIDE.htm
http://www.bing.com
http://www.find.org.tw
http://www.google.com
http://www.vision.caltech.edu/Image_Datasets/Caltech256/