Abstract To reduce the gap between low-level image features and high-level semantic concept, support vector machine based relevance feedback draws more and more attentions. However, the information embedded in unlabeled samples is not utilized in that method. In order to exploit these information sufficiently, the transductive support vector machine (TSVM) is introduced into feedback process. Based on analyzing the characters of feature vector for TSVM, a color sparse feature is designed as the image description feature combined with the texture feature. Experimental results show that the proposed method is more discriminative than the feedback process using support vector machine (SVM), and TSVM obtains good results when applied to other fields.
[1] Rui Yong, Huang T S, Ortega M, et al. Relevance Feedback: A Power Tool in Interactive Content-Based Image Retrieval. IEEE Trans on Circuits and System for Video Technology, 1998, 8(5): 644-655 [2] Ma Chao, Tang Zhide. Application of Relevance Feedback Techniques in Content-Based Image Retrieval. Journal of Chongqing University of Science and Technology: Natural Sciences Edition, 2007, 9(1): 81-84 (in Chinese) (马 超,唐治德. 相关反馈技术在图像检索系统中的应用. 重庆科技学院学报:自然科学版,2007, 9(1): 81-84) [3] Vapnik V N. The Nature of Statistical Learning Theory. New York, USA: Springer-Verlag, 1995 [4] Zhou Zhihua, Chen Keijia, Jiang Yuan. Exploiting Unlabeled Data in Content-Based Image Retrieval // Proc of the 15th European Conference on Machine Learning. Pisa, Italy, 2004: 525-536 [5] Qiu Bo, Xu Changsheng, Tian Qi. Efficient Relevance Feedback Using Semi-Supervised Kernel-Specified K-means Clustering // Proc of the 18th International Conference on Pattern Recognition. Hongkong, China, 2006, Ⅲ: 316-319 [6] Joachims T. Transductive Inference for Text Classification Using Support Vector Machines // Proc of the 16th International Conference on Machine Learning. San Francisco, USA, 1999: 200-209 [7] Vapnik V N. Statistical Learning Theory. New York, USA: Wiley, 1998 [8] Wan Hualin, Chowdhury M U. Image Semantic Classification by Using SVM. Journal of Software, 2003, 14(11): 1891-1899 (in Chinese) (万华林,Chowdhury M U. 基于支持向量机的图像语义分类.软件学报, 2003, 14(11): 1891-1899) [9] Wang Ye, Huang S T. Training TSVM with the Proper Number of Positive Samples. Pattern Recognition Letter, 2005, 26(14): 2187-2194 [10] Bennett K P, Demiriz A. Semi-Supervised Support Vector Machines [EB/OL]. [1998-07-21]. http://www1.cs.columbia.edu/~dplewis/candidacy/bennett98semisupervised.pdf [11] Chen Yisong, Wang Guopin, Dong Shihai. A Progressive Transductive Inference Algorithm Based on Support Vector Machine. Journal of Software, 2003, 14(3): 451-460 (in Chinese) (陈毅松,汪国平,董士海.基于支持向量机的渐进直推式分类学习算法.软件学报, 2003, 14(3): 451-460) [12] Zhong Qingliu, Cai Zixing. Semi-Supervised Learning Algorithm Based on SVM and by Gradual Approach. Computer Engineering and Applications, 2006, 42(15): 19-22 (in Chinese) (钟清流,蔡自兴.基于支持向量机的渐近式半监督式学习算法.计算机工程与应用, 2006, 42(15): 19-22) [13] Zhang Rong, Alexander I. A New Data Selection Principle for Semi-Supervised Incremental Learning // Proc of the 18th International Conference on Pattern Recognition. Hongkong, China, 2006: 780-783 [14] Wang J Z, Li Jia. Learning-Based Linguistic Indexing of Pictures with 2D MHMMs // Proc of the 10th ACM Conference on Multimedia. Juan-Les-Pins, France, 2002: 436-445 [15] Wu D H, Bennett K P, Cristianini N, et al. Large Margin Decision Trees for Induction and Transduction // Proc of the 16th International Conference of Machine Learning. Bled, Slovenia, 1999: 474-483 [16] Joachims T. Text Categorization with Support Vector Machine: Learning with Many Relevant Features // Proc of the European Conference on Machine Learning. Chemnitz, Germany, 1998: 137-142 [17] Donoho D L, Elad M. Optimally Sparse Representation in General (Nonorthogonal) Dictionaries via l1 Minimization. Proc of the National Academy of Sciences of the United States of America, 2003, 100(5): 2197-2202 [18] Kivinen J, Warmuth M. The Perceptron Algorithm vs. Winnow: Linear vs. Logarithmic Mistake Bounds When Few Input Variables Are Relevant. Santa Cruz, USA: University of California Press, 1995 [19] Huang Jing, Kumar S R, Mitra M, et al. Image Indexing Using Color Correlograms // Proc of the IEEE Conference on Computer Vision and Pattern Recognition. San Juan, Puerto Rico, 1997: 762-768
2009, Vol. 22 
