Abstract
Image representation has been a fundamental problem for many real world applications, such as image database visualization, browsing, retrieval, etc. In this paper, we investigate the use of Laplacian Eigenmap (LE) for image representation and retrieval. Conventional, Principal Component Analysis (PCA) has been considered effective as to discovering the low dimensional structure of the image space. However, PCA can only discover the linear structure. It fails when the images are sampled from a low dimensional nonlinear manifold which is embedded in the high dimensional Euclidean space. By using Laplacian Eigenmap, we first build a nearest neighbor graph which models the local geometrical structure of the image space. A locality preserving mapping is then obtained to respect the graph structure. We compared the PCA and LE based image representations in the context of image retrieval. Experimental results show the effectiveness of the LE based representation.
Access this chapter
Tax calculation will be finalised at checkout
Purchases are for personal use only
Preview
Unable to display preview. Download preview PDF.
References
Belkin, M., Niyogi, P.: Laplacian Eigenmaps and Spectral Techniques for Embedding and Clustering. In: Advances in Neural Information Processing Systems 14, Vancouver, Canada (2001)
He, X., King, O., Ma, W.-Y., Li, M., Zhang, H.-J.: Learning a Semantic Space from User’s Relevance Feedback for Image Retrieval. IEEE Trans. On Circuit and Systems for Video Technology 13(1) (2003)
Rui, Y., Huang, T.S., Chang, S.-F.: Image Retrieval: Current Techniques, Promising Directions and Open Issues. Journal of Visual Communication and Image Representation 10, 39–62 (1999)
Author information
Authors and Affiliations
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 2004 Springer-Verlag Berlin Heidelberg
About this paper
Cite this paper
Lu, K., He, X., Zeng, J. (2004). Image Retrieval Using Dimensionality Reduction. In: Zhang, J., He, JH., Fu, Y. (eds) Computational and Information Science. CIS 2004. Lecture Notes in Computer Science, vol 3314. Springer, Berlin, Heidelberg. https://doi.org/10.1007/978-3-540-30497-5_120
Download citation
DOI: https://doi.org/10.1007/978-3-540-30497-5_120
Publisher Name: Springer, Berlin, Heidelberg
Print ISBN: 978-3-540-24127-0
Online ISBN: 978-3-540-30497-5
eBook Packages: Computer ScienceComputer Science (R0)