Support Vector Machine Learning for Region-Based Image Retrieval with Relevance Feedback

ETRI Journal

关键词: region-based image retrieval;    relevance feedback;    cluster-merging;    Support vector machine;   

Others  :  1185527
DOI  :  10.4218/etrij.07.0207.0037

【 摘 要 】

We present a relevance feedback approach based on multi-class support vector machine (SVM) learning and cluster-merging which can significantly improve the retrieval performance in region-based image retrieval. Semantically relevant images may exhibit various visual characteristics and may be scattered in several classes in the feature space due to the semantic gap between low-level features and high-level semantics in the user's mind. To find the semantic classes through relevance feedback, the proposed method reduces the burden of completely re-clustering the classes at iterations and classifies multiple classes. Experimental results show that the proposed method is more effective and efficient than the two-class SVM and multi-class relevance feedback methods.

   

【 参考文献 】

• [1]L. Zhang, F. Lin, and B. Zhang, "Support Vector Machine Learning for Image Retrieval," Proc. IEEE ICIP, 2001.
• [2]F. Jing, M. Li, H.-J. Zhang, and B. Zhang, "Support Vector Machines for Region-Based Image Retrieval," Proc. Int’l Conf. on Multimedia and Expo, vol. 1, 2003, pp. 21-24.
• [3]J. Peng, "Multi-Class Relevance Feedback Content-Based Image Retrieval," Computer Vision and Image Understanding, vol. 90, no. 1, 2003, pp. 42-67.
• [4]D.-H. Kim and S.-L.Lee, "Relevance Feedback Using Adaptive Clustering for Region Based Image Similarity Retrieval," LNAI 4099, 2006, pp. 641-650.