Regularized directional feature learning for face recognition

Autor:	Demetrio Labate, Mohamed Anouar Borgi, Maher El'arbi, Chokri Ben Amar
Rok vydání:	2014
Předmět:	Biometrics Computer Networks and Communications Computer science business.industry Pattern recognition Sparse approximation Facial recognition system Sample (graphics) Hardware and Architecture Shearlet Media Technology Computer vision Artificial intelligence business Feature learning Software
Zdroj:	Multimedia Tools and Applications. 74:11281-11295
ISSN:	1573-7721 1380-7501
DOI:	10.1007/s11042-014-2228-3
Popis:	This paper presents an improved approach to face recognition, called Regularized Shearlet Network (RSN), which takes advantage of the sparse representation properties of shearlets in biometric applications. One of the novelties of our approach is that directional and anisotropic geometric features are efficiently extracted and used for the recognition step. In addition, our approach is augmented by regularization theory (RSN) in order to control the trade-off between the fidelity to the data (gallery) and the smoothness of the solution (probe). In this work, we address the challenging problem of the single training sample per subject (STSS). We compare our new algorithm against different state-of-the-art methods.
Databáze:	OpenAIRE
Externí odkaz:	https://explore.openaire.eu/search/publication?articleId=doi_________::a56f11dce0da32bc15e549f2651337e8 https://doi.org/10.1007/s11042-014-2228-3 Zobrazit plný text záznamu Full text from SpringerLink