Single-sample face recognition under varying lighting conditions based on logarithmic total variation

Zhang, Y; Lu, X; Li, J

Single-sample face recognition under varying lighting conditions based on logarithmic total variation

Zhang, Y Lu, X Li, J

Permalink

Publication Type:: Journal Article
Citation:: Signal, Image and Video Processing, 2019, 13 (4), pp. 657 - 665
Issue Date:: 2019-06-01

Closed Access

	Filename	Description	Size
	Zhang2019_Article_Single-sampleFaceRecognitionUn.pdf	Published Version	1.5 MB	Adobe PDF	View/Open

Copyright Clearance Process

Recently Added
In Progress
Closed Access

This item is closed access and not available.

Full metadata record

Field	Value	Language
dc.contributor.author	Zhang, Y	en_US
dc.contributor.author	Lu, X	en_US
dc.contributor.author	Li, J https://orcid.org/0000-0002-1336-2241	en_US
dc.date.accessioned	2020-04-03T04:51:44Z
dc.date.available	2020-04-03T04:51:44Z
dc.date.issued	2019-06-01	en_US
dc.identifier.citation	Signal, Image and Video Processing, 2019, 13 (4), pp. 657 - 665	en_US
dc.identifier.issn	1863-1703	en_US
dc.identifier.uri	http://hdl.handle.net/10453/139772
dc.description.abstract	© 2018, Springer-Verlag London Ltd., part of Springer Nature. The logarithmic total variation (LTV) algorithm is a classical algorithm that is proposed to address the illumination interference in face recognition. Some state-of-the-art techniques based on LTV assume that the illumination component mainly lies in the low-frequency features among face images. However, these techniques adopt unsuitable methods to process low-frequency features, resulting in final unsatisfactory recognition rates. In this paper, we propose an improved illumination normalization method based on the LTV method, called the RETINA&TH-LTV algorithm. In this algorithm, the retina model is utilized to eliminate most of the illumination component in low-frequency features. Then, an advanced contrast-limited adaptive histogram equalization technique is proposed to remove the residual lighting component. At the same time, through realizing threshold-value filtering on high-frequency features, the enhancement of facial features is achieved. Finally, the processed frequency features are combined to form a robust holistic feature image, which is then utilized for recognition. Insufficient training images in face recognition are also taken into consideration in this research. Comparative experiments for single-sample face recognition are conducted on YALE B, CMU PIE and our self-built driver databases. The nearest neighbor classifier and extended sparse representation classifier are employed as classification methods. The results indicate that the RETINA&TH-LTV algorithm has promising performance, especially in serious illumination and insufficient training sample conditions.	en_US
dc.relation.ispartof	Signal, Image and Video Processing	en_US
dc.relation.isbasedon	10.1007/s11760-018-1394-6	en_US
dc.rights	info:eu-repo/semantics/closedAccess
dc.subject.classification	Artificial Intelligence & Image Processing	en_US
dc.title	Single-sample face recognition under varying lighting conditions based on logarithmic total variation	en_US
dc.type	Journal Article
utslib.citation.volume	4	en_US
utslib.citation.volume	13	en_US
utslib.for	0801 Artificial Intelligence and Image Processing	en_US
utslib.for	0906 Electrical and Electronic Engineering	en_US
pubs.embargo.period	Not known	en_US
pubs.organisational-group	/University of Technology Sydney
pubs.organisational-group	/University of Technology Sydney/Faculty of Engineering and Information Technology
pubs.organisational-group	/University of Technology Sydney/Faculty of Engineering and Information Technology/School of Computer Science
pubs.organisational-group	/University of Technology Sydney/Strength - CAI - Centre for Artificial Intelligence
utslib.copyright.status	closed_access	*
pubs.issue	4	en_US
pubs.publication-status	Published	en_US
pubs.volume	13	en_US

Abstract:

© 2018, Springer-Verlag London Ltd., part of Springer Nature. The logarithmic total variation (LTV) algorithm is a classical algorithm that is proposed to address the illumination interference in face recognition. Some state-of-the-art techniques based on LTV assume that the illumination component mainly lies in the low-frequency features among face images. However, these techniques adopt unsuitable methods to process low-frequency features, resulting in final unsatisfactory recognition rates. In this paper, we propose an improved illumination normalization method based on the LTV method, called the RETINA&TH-LTV algorithm. In this algorithm, the retina model is utilized to eliminate most of the illumination component in low-frequency features. Then, an advanced contrast-limited adaptive histogram equalization technique is proposed to remove the residual lighting component. At the same time, through realizing threshold-value filtering on high-frequency features, the enhancement of facial features is achieved. Finally, the processed frequency features are combined to form a robust holistic feature image, which is then utilized for recognition. Insufficient training images in face recognition are also taken into consideration in this research. Comparative experiments for single-sample face recognition are conducted on YALE B, CMU PIE and our self-built driver databases. The nearest neighbor classifier and extended sparse representation classifier are employed as classification methods. The results indicate that the RETINA&TH-LTV algorithm has promising performance, especially in serious illumination and insufficient training sample conditions.

Please use this identifier to cite or link to this item:

http://hdl.handle.net/10453/139772