Abstract
There is a great deal of research conducted on hyperplane based query such as Support Vector Machine (SVM) in Content-based Image Retrieval(CBIR). However, the SVM-based CBIR always suffers from the problem of the imbalance of image data. Specifically, the number of negative samples (irrelevant images) is far more than that of the positive ones. To deal with this problem, we propose a new active learning approach to enhance the positive sample set in SVM-based Web image retrieval. In our method, instead of using complex parsing methods to analyze Web pages, two kinds of “lightweight” image features: the URL of the Web image and its visual features, which can be easily obtained, are applied to estimate the probability of the image being a potential positive sample. The experiments conducted on a test data set with more than 10,000 images from about 50 different Web sites demonstrate that compared with traditional methods, our approach improves the retrieval performance significantly.
This is a preview of subscription content, log in via an institution to check access.
Preview
Unable to display preview. Download preview PDF.
Similar content being viewed by others
References
Rui, Y., Huang, T., Ortega, M., Mehrotra, S.: Relevance feedback:A power tool in interactive content-based image retrieval. IEEE Tran. on Circuits and Systems for Video Technology 8(5) (1998)
Burges, C.: A Tutorial On Support Vector Machines For Pattern Recognition. Data mining and Knowledge Discovery (1998)
Chang, E.Y., Lai, W.-C.: Active Learning and its Scalability for Image Retrieval. In: IEEE ICME (2004)
Brinker, K.: Incorporating diversity in active learning with support vector machines. In: ICML (2003)
Tong, S., Chang, E.: Support vector machine active learning for image retrieval. In: ACM MM 2001 (2001)
Chen, Y., Zhou, X., Huang, T.: One-class SVM For Learning In Image Retrieval. In: IEEE ICIP 2001, Thessaloniki, Greece (2001)
Gosselin, P.H., Cord, M.: Active Learning Techniques for User Interactive Systems: Application to Image Retrieval, Machine Learning Techniques for Processing Multimedia Content, Bonn, Germany (2005)
Cai, D., Xiaofei,: Hierarchical Clustering of WWW Image Search Results Using Visual, Textual and Link Information. In: ACM MM 2004 (2004)
Goh, K.S., Chang, E., Lai, W.C.: Multimodal Concept-Dependeng Active Learning for Image Retrieval. In: ACM MM 2004 (2004)
Quack, T., Monich, U., Thiele, L., Manjunath, B.S.: Cortina: A System for Large-scale, Content-based Web Image Retrieval. In: ACM MM 2004 (2004)
Jing, F., Li, M., Zhang, H.J., Zhang, B.: Support Vector Machines for Region-Based Image Retrieval. In: IEEE ICME (2003)
Huang, T.S., Zhou, X.S.: Image retrieval by relevance feedback:from heuristic weight adjustment to optimal learning methods. In: IEEE ICIP (2001)
He, X., Ma, W.Y., Zhang, H.-J.: ImageSeer:Clustering and Searching WWW Images Using Link and Page Layout Analysis, Micsoft Technical Report (2004)
Hua, Z., Wang, X.J., Liu, Q.: Semantic knowledge Extraction and Annotation for Web Images. In: ACM MM 2005 (2005)
Author information
Authors and Affiliations
Editor information
Rights and permissions
Copyright information
© 2007 Springer-Verlag Berlin Heidelberg
About this paper
Cite this paper
Yuan, J., Zhou, X., Xu, H., Wang, M., Wang, W. (2007). A Novel Active Learning Approach for SVM Based Web Image Retrieval. In: Ip, H.HS., Au, O.C., Leung, H., Sun, MT., Ma, WY., Hu, SM. (eds) Advances in Multimedia Information Processing – PCM 2007. PCM 2007. Lecture Notes in Computer Science, vol 4810. Springer, Berlin, Heidelberg. https://doi.org/10.1007/978-3-540-77255-2_5
Download citation
DOI: https://doi.org/10.1007/978-3-540-77255-2_5
Publisher Name: Springer, Berlin, Heidelberg
Print ISBN: 978-3-540-77254-5
Online ISBN: 978-3-540-77255-2
eBook Packages: Computer ScienceComputer Science (R0)