基于Nystr&ouml;m 低阶近似的半监督流形排序图像检索

傅向华; 李坚强; 王志强; 杜文峰

doi:10.3724/SP.J.1004.2011.00787

基于Nyström 低阶近似的半监督流形排序图像检索

doi: 10.3724/SP.J.1004.2011.00787

1.
深圳大学计算机与软件学院深圳 518060

计量
- 文章访问数: 2844
- HTML全文浏览量: 42
- PDF下载量: 1000
- 被引次数: 0
出版历程
- 收稿日期: 2010-01-21
- 修回日期: 2010-11-24
- 刊出日期: 2011-07-20

Semi-supervised Manifold-ranking-based Image Retrieval with Low-rank Nyström Approximation

1.
College of Computer Science and Software Engineering, Shenzhen University, Shenzhen 518060

摘要

摘要: 实际图像检索过程中,用户提供的相关反馈有限,但存在大量未标记图像数据. 本文在前期半监督流形图像检索工作的基础上,提出一种基于Nystrm低阶近似的半监督流形排序图像检索方法.通过采用半监督的流形正则化框架, 将图像数据嵌入到低维流形结构中进行分类排序,以充分利用大量未标记数据, 并兼顾分类误差、数据分布的几何结构以及分类函数的复杂性.针对半监督学习速度缓慢的问题, 基于Nystrm低阶近似对学习过程进行加速.在较大规模的Corel图像数据集上进行了检索实验, 实验结果表明该方法能获得较好的效果.
- 图像检索 /
- 流形学习 /
- Nystr? /
- m近似 /
- 半监督学习
Abstract: In the real image retrieval process, there are abundant unlabeled images whereas there only exist few labeled images. To address this issue, based on our previous work of semi-supervised manifold image retrieval, this paper proposed a novel learning method named semi-supervised manifold ranking based image retrieval (S2MRBIR). The images are assumed always embedded in low-dimensional sub-manifolds. In particular, S2MRBIR adopts the manifold regularization framework to rank the retrieved images while regarding the relevant feedback process of image retrieval as an online learning process and treating the image retrieval as a classification problem. The manifold regularization framework is capable of taking account of both labeled and unlabeled data, the classification performance, the geometric structures of the data distribution, and the complexity of the classifier. Moreover, an accelerating algorithm based on Low-rank Nystrm approximation was proposed to improve the computing procedure of S2MRBIR (NA-S2MRBIR). Experimental results on Corel image database demonstrated the effectiveness of S2MRBIR.
- Image retrieval /
- manifold learning /
- Nyströ /
- m approximation /
- semi-supervised learning

HTML全文

参考文献(1)

[1]

Smeulders A W M, Worring M, Santini S, Gupta A, Jain R. Content-based image retrieval at the end of the early years. IEEE Transactions on Pattern Analysis and Machine Intelligence, 2000, 22(12): 1349-1380[2] Liu Qing, Xu Lu-Ping, Ma Yi-De, Wang Yong. Image NMI feature extraction and retrieval method based on pulse coupled neural networks. Acta Automatica Sinica, 2010, 36(7): 931-938(刘勍, 许录平, 马义德, 王勇. 基于脉冲耦合神经网络的图像NMI特征提取及检索方法. 自动化学报, 2010, 36(7): 931-938[3] Liu Y, Zhang D S, Lu G J, Ma W Y. A survey of content-based image retrieval with high-level semantics. Pattern Recognition, 2007, 40(1): 262-282 [4] Arevalillo-Herraez M, Ferri F J, Domingo J. A naive relevance feedback model for content-based image retrieval using multiple similarity measures. Pattern Recognition, 43(3): 619-629[5] Zhou Z H, Chen K J, Dai H B. Enhancing relevance feedback in image retrieval using unlabeled data. ACM Transactions on Information Systems, 2006, 24(2): 219-244[6] Luo Si-Wei, Zhao Lian-Wei. Manifold learning algorithms based on spectral graph theory. Journal of Computer Research and Development, 2006, 43(7): 1173-1179(罗四维, 赵连伟. 基于谱图理论的流形学习算法. 计算机研究与发展, 2006, 43(7): 1173-1179)[7] Zhou D Y, Weston J, Gretton A, Bousquet O, Scholkopf B. Ranking on data manifolds. Advances in Neural Information Processing Systems. Cambridge: The MIT Press, 2004. 169-176[8] He J R, Li M J, Zhang H J, Tong H H, Zhang C S. Generalized manifold-ranking-based image retrieval. IEEE Transactions on Image Processing, 2006, 15(10): 3170-3177[9] Wang C, Zhao J, He X F, Chen C, Bu J J. Image retrieval using nonlinear manifold embedding. Neurocomputing, 2009, 72(16-18): 3922-3929[10] Lin Y Y, Liu T L, Chen H T. Semantic manifold learning for image retrieval. In: Proceedings of the 13th Annual ACM International Conference on Multimedia. New York, USA: ACM, 2005. 249-258[11] Li Jie, Cheng Yi-Min, Ge Shi-Ming, Zhang Ling. Integrated manifold ranking and region matching for image retrieval. Journal of Chinese Computer Systems, 2008, 29(3): 511-515(李杰, 程义民, 葛仕明, 张玲. 结合流形排序和区域匹配的图像检索. 小型微型计算机系统, 2008, 29(3): 511-515)[12] Wang Zhi-Qiang, Fu Xiang-Hua, Zhao Liang-Hui, Du Wen-Feng. Manifold based semi-supervised learning for content-based image retrieval. Geomatics and Information Science of Wuhan University, 2009, 34(8): 928-931, 935(王志强, 傅向华, 赵良辉, 杜文峰. 基于内容的半监督流体图像检索. 武汉大学学报信息科学版, 2009, 34(8): 928-931, 935)[13] Belkin M, Niyogi P, Sindhwani V. Manifold regularization: a geometric framework for learning from labeled and unlabeled examples. The Journal of Machine Learning Research, 2006, 7: 2399-2434[14] Zhang K, Tsang I W, Kwok J T. Improved Nystrm low-rank approximation and error analysis. In: Proceedings of the 25th International Conference on Machine Learning. New York, USA: ACM, 2008. 1232-1239[15] Huijsmans D P, Sebe N. How to complete performance graphs in content-based image retrieval: add generality and normalize scope. IEEE Transactions on Pattern Analysis and Machine Intelligence, 2005, 27(2): 245-251

施引文献

资源附件(0)

访问统计