Multilinear Principal Component Analysis Network for Tensor Object Classification

Abstract : The recently proposed principal component analysis network (PCANet) has performed well with respect to the classification of 2-D images. However, feature extraction may perform less well when dealing with multi-dimensional images, since the spatial relationships within the structures of the images are not fully utilized. In this paper, we develop a multilinear principal component analysis network (MPCANet), which is a tensor extension of PCANet, to extract the high-level semantic features from multi-dimensional images. The extracted features largely minimize the intraclass invariance of tensor objects by making efficient use of spatial relationships within multi-dimensional images. The proposed MPCANet outperforms traditional methods on a benchmark composed of three data sets, including the UCF sports action database, the UCF11 database, and a medical image database. It is shown that even a simple one-layer MPCANet may outperform a two-layer PCANet. © 2017 IEEE.
Liste complète des métadonnées

https://hal-univ-rennes1.archives-ouvertes.fr/hal-01618929
Contributeur : Laurent Jonchère <>
Soumis le : mercredi 18 octobre 2017 - 17:15:11
Dernière modification le : mardi 19 décembre 2017 - 10:48:04

Identifiants

Collections

Citation

J. Wu, S. Qiu, R. Zeng, Y. Kong, Lotfi Senhadji, et al.. Multilinear Principal Component Analysis Network for Tensor Object Classification. IEEE Access, 2017, 5, pp.3322-3331. 〈10.1109/ACCESS.2017.2675478〉. 〈hal-01618929〉

Partager

Métriques

Consultations de la notice

26