Abstract An approach is proposed for keyword-based image retrieval in unannotated image databases. After the keyword is input, the surrounding text information of the images is used to filter some irrelevant images, and then the visual information is extracted to select the relevant images. The data editing techniques are employed to refine the relevant images which are used as queries for retrieving images from the image databases. Experimental results show that the proposed method can achieve good retrieval performance in unannotated image databases.
[1] Smeulders A W M, Worring M, Santini S, et al. Content-Based Image Retrieval at the End of the Early Years. IEEE Trans on Pattern Analysis and Machine Intelligence, 2000, 22(12): 1349-1380 [2] Hoi C H, Lyu M R. Web Image Learning for Searching Semantic Concepts in Image Databases // Proc of the 13th International World Wide Web Conference on Alternate Track Papers and Posters. New York, USA, 2004: 406-407 [3] Yanai K.Generic Image Classification Using Visual Knowledge on the Web // Proc of the 11th ACM International Conference on Multimedia. Berkeley, USA, 2003: 167-176 [4] Barandela R, Gasca E. Decontamination of Training Samples for Supervised Pattern Recognition Methods // Proc of the Joint IAPR International Workshops on Advances in Pattern Recognition. Alicante, Spain, 2000: 621-630 [5] Jiang Yuan, Zhou Zhihua. Editing Training Data for kNN Classifiers with Neural Network Ensemble // Proc of the 1st International Symposium on Neural Networks. Dalian, China, 2004: 356-361 [6] Li Ming, Zhou Zhihua. SETRED: Self-Training with Editing // Proc of the 9th Pacific-Asia Conference on Knowledge Discovery and Data Mining. Hanoi, Vietnam, 2005: 611-621 [7] Zhou Zhihua, Dai Hongbin. Exploiting Image Contents in Web Search // Proc of the 20th International Joint Conferences on Artificial Intelligence. Hyderabad, India, 2007: 2928-2933
2009, Vol. 22 
