Classification of scenes based on multiway feature extraction Phan A.H., Cichocki A., Vu-Dinh T. Lab for Advanced Brain Signal Processing, Brain Science Institute, RIKEN, Japan; Hochiminh City University of Technology, Hochiminh City, Viet Nam Abstract: Recognition of real world scenes can be efficiently solved based on global features termed the Spatial Envelope. Such features indeed comprise multiple modes such as orientations, scales, sparsity profiles. In order to extract features and classify multiway samples, most approaches vectorize data tensors to convert the classification of multiway data into the one of 1-D samples. This common approach disregards the multiway structures of global features, hence it can face the risk of losing correlation information between modes (orientations or scales). To this end, by revisiting the problem of scene classification in view of tensor decompositions, a new method is introduced to extract multiway features. The projection filter is designed for global features based on a set of basis matrices instead of only one basis as in 1-D problem. The proposed approach not only improves the classification accuracy, but also reduces the running time for training stage and feature projection. ©2010 IEEE. Index Keywords: Classification accuracy; Feature projection; Global feature; Multiple modes; Running time; Scene classification; Tensor decomposition; Data handling; Tensors; Feature extraction Year: 2010 Source title: Proceedings - 2010 International Conference on Advanced Technologies for Communications, ATC 2010 Art. No.: 5672694 Page : 142-145 Link: Scorpus Link Correspondence Address: Phan, A. H.; Lab for Advanced Brain Signal Processing, Brain Science Institute, RIKENJapan; email:
[email protected] Sponsors: IEEE;IEEE Communications Society;REV;VNU HCMC;ICT Conference name: 2010 International Conference on Advanced Technologies for Communications, ATC 2010 Conference date: 20 October 2010 through 22 October 2010 Conference location: Ho Chi Minh City Conference code: 83502 ISBN: 9.78142E+12 DOI: 10.1109/ATC.2010.5672694 Language of Original Document: English Abbreviated Source Title: Proceedings - 2010 International Conference on Advanced Technologies for Communications, ATC 2010 Document Type: Conference Paper
Source: Scopus Authors with affiliations: 1. Phan, A.H., Lab for Advanced Brain Signal Processing, Brain Science Institute, RIKEN, Japan 2. Cichocki, A., Lab for Advanced Brain Signal Processing, Brain Science Institute, RIKEN, Japan 3. Vu-Dinh, T., Hochiminh City University of Technology, Hochiminh City, Viet Nam
References: 1.
Biederman, I., Aspects and extension of a theory of human image understanding (1988) Computational Process in Human Vision: An Interdisciplinary Perspective
2.
Oliva, A., Torralba, A., Modeling the shape of the scene: A holistic representation of the spatial envelope (2001) International Journal of Computer Vision, 42 (3), pp. 145-175
3.
Torralba, A., Oliva, A., Semantic organization of scenes using discriminant structural templates In International Conference on Computer Vision ICCV99, 1999, pp. 1253-1258
4.
Lee, T.S., Image representation using 2d Gabor wavelets (1996) IEEE Trans. Pattern Analysis and Machine Intelligence, 18, pp. 959-971
5.
Cichocki, A., Zdunek, R., Phan, A.-H., Amari, S., (2009) Nonnegative Matrix and Tensor Factorizations: Applications to Exploratory Multi-way Data Analysis and Blind Source Separation, , Chichester: Wiley
6.
Tucker, L., Some mathematical notes on three-mode factor analysis (1966) Psychometrika, 31, pp. 279-311
7.
Canu, S., Grandvalet, Y., Guigue, V., Rakotomamonjy, A., SVM and kernel methods -Matlab toolbox (2005) Perception Systmes et Information, , INSA de Rouen, Rouen, France
8.
Phan, A.-H., Cichocki, A., Fast and efficient algorithms for nonnegative Tucker decomposition (2008) LNCS, 5264, pp. 772782. , Proc. of The Fifth International Symposium on Neural Networks, Springer Beijing, China, 24-28, September
9.
Phan, A.-H., Cichocki, A., Local learning rules for nonnegative Tucker decomposition (2009) ICONIP
10. Phan, A.-H., Cichocki, A., Extended HALS algorithm for nonnegative Tucker decomposition and applications for classification and analysis of multidimensional data (2010) Neurocomputing, , (accepted) 11. Phan, A.-H., Cichocki, A., Tensor decompositions for feature extraction and classification of high dimensional datasets (2010) Nonlinear Theory and Its Applications, IEICE, , (invited paper, submitted)