SPECIAL ISSUE: Fusion of Multiple Classifiers; Edited by F. Roli and J. KittlerNo Access

A STUDY ON COMBINING IMAGE REPRESENTATIONS FOR IMAGE CLASSIFICATION AND RETRIEVAL

CARMEN LAI

Pattern Recognition Group, TU Delft, Lorentzweg 1, Delft, 2628 CJ, The Netherlands

Search for more papers by this author

DAVID M. J. TAX

Fraunhofer FIRST.IDA, Kekuléstr.7, Berlin, D-12489, Germany

Search for more papers by this author

ROBERT P. W. DUIN

Pattern Recognition Group, TU Delft, Lorentzweg 1, Delft, 2628 CJ, The Netherlands

Search for more papers by this author

ELŻBIETA PĘKALSKA

Pattern Recognition Group, TU Delft, Lorentzweg 1, Delft, 2628 CJ, The Netherlands

Search for more papers by this author

, and

PAVEL PACLÍK

Pattern Recognition Group, TU Delft, Lorentzweg 1, Delft, 2628 CJ, The Netherlands

Search for more papers by this author

https://doi.org/10.1142/S0218001404003459Cited by:30 (Source: Crossref)

Abstract

A flexible description of images is offered by a cloud of points in a feature space. In the context of image retrieval such clouds can be represented in a number of ways. Two approaches are here considered. The first approach is based on the assumption of a normal distribution, hence homogeneous clouds, while the second one focuses on the boundary description, which is more suitable for multimodal clouds. The images are then compared either by using the Mahalanobis distance or by the support vector data description (SVDD), respectively.

The paper investigates some possibilities of combining the image clouds based on the idea that responses of several cloud descriptions may convey a pattern, specific for semantically similar images. A ranking of image dissimilarities is used as a comparison for two image databases targeting image classification and retrieval problems. We show that combining of the SVDD descriptions improves the retrieval performance with respect to ranking, on the contrary to the Mahalanobis case. Surprisingly, it turns out that the ranking of the Mahalanobis distances works well also for inhomogeneous images.

The final work on this paper was done at the Information and Communication Theory Group, TU Delft, P.O. Box 5031, 2600 GA, Delft, Netherlands.

Keywords:

References

S. Antani, R. Kasturi and R. Jain, Pattern recognition methods in image and video databases: past, present and future, Advances in Pattern Recognition, Proc. SPR'98 and SSPR'98 (Springer-Verlag, Berlin, 1998) pp. 31–53. Google Scholar
C. M. Bishop , Neural Networks for Pattern Recognition ( Oxford University Press , 1995 ) . Crossref, Google Scholar
T. Gevers, F. Aldershoff and A. W. M. Smeulders, Classification of images on internet by visual and textual information, Internet Imaging, SPIE 3964, eds. G. B. Beretta and R. Schettini (2000) pp. 16–27. Google Scholar
T. Huang , Y. Rui and S.-F. Chang , Image retrieval: past, present, and future , Int. Symp. Multimedia Information Processing ( 1997 ) . Google Scholar
L. I. Kuncheva, J. C. Bezdek and R. P. W. Duin, Patt. Recogn. 34(2), 299 (2001), DOI: 10.1016/S0031-3203(99)00223-X. Crossref, Web of Science, Google Scholar
C. Lai et al. , On combining one-class classifiers for image database retrieval , Multiple Classifier Systems, Proc. Third International Workshop MCS 2002 , Lecture Notes in Computer Science 2364 , eds. F. Roli and J. Kittler ( Springer-Verlag , Berlin , 2002 ) . Google Scholar
M. S. Lew , Principles of Visual Information Retrieval ( Springer-Verlag , London , 2001 ) . Crossref, Google Scholar
O. Maron and A. L. Ratan, Multiple-instance learning for natural scene classification, Proc. 15th Int. Conf. Machine Learning (Morgan Kaufmann, San Francisco, CA, 1998) pp. 341–349. Google Scholar
K. Messer, Automatic image database retrieval system using adaptive colour and texture descriptors, Ph.D. thesis, University of Surrey, Guildford, 1999 . Google Scholar
K. Messer and J. Kittler, Patt. Recogn. Lett. 20, 1323 (1999), DOI: 10.1016/S0167-8655(99)00101-4. Crossref, Web of Science, Google Scholar
El. Pękalska and R. P. W. Duin, Patt. Recogn. Lett. 23(8), 943 (2002), DOI: 10.1016/S0167-8655(02)00024-7. Crossref, Web of Science, Google Scholar
Y. Rui, T. Huang and S. Mehrotra, Content-based image retrieval with relevance feedback in MARS, Proc. IEEE Int. Conf. Image Proc. (1997) pp. 815–818. Google Scholar
B. Schölkopf et al. , Advances in Neural Information Processing Systems , eds. M. S. Kearns , S. A. Solla and D. A. Cohn ( 1999 ) . Google Scholar
A. Smeulderset al., IEEE Trans. Patt. Anal. Mach. Intel. 22(12), 1349 (2000), DOI: 10.1109/34.895972. Crossref, Web of Science, Google Scholar
D. M. J. Tax, One-class classification, Ph.D. thesis, Delft University of Technology, http://www.ph.tn.tudelft.nl/~davidt/thesis.pdf, June 2001 . Google Scholar
D. M. J. Tax and R. P. W. Duin, Patt. Recogn. Lett. 20(11–13), 1191 (1999), DOI: 10.1016/S0167-8655(99)00087-2. Crossref, Web of Science, Google Scholar
V. N. Vapnik , Statistical Learning Theory ( John Wiley , 1998 ) . Google Scholar