The Impact of Image Quality on the Performance of Face Recognition

  • 8 Citations

Abstract

The performance of a face recognition system depends on the quality of both test and reference images participating in the face comparison process. In a forensic evaluation case involving face recognition, we do not have any control over the quality of the trace (image captured by a CCTV at a crime scene) and it is usually of very low quality. However, forensic investigators have some control over the quality of suspects reference images. In this paper, we investigate if it is useful to modify the quality of the reference (usually, good quality mug shot) images in order to achieve better recognition performance using a view based face recognition system. We found that approximately matching non-frontal pose between test and reference images can greatly improve recognition performance. Moreover, it is the relative difference in pose between the test and reference image that determines the extent of influence that other quality parameters like illumination, noise, motion blur, and resolution have on the face recognition performance.
Original languageUndefined
Title of host publication33rd WIC Symposium on Information Theory in the Benelux
Place of PublicationEnschede, the Netherlands
PublisherCentre for Telematics and Information Technology (CTIT)
Pages141-148
Number of pages8
ISBN (Print)978-90-365-3383-6
StatePublished - May 2012

Publication series

Name
PublisherUniversity of Twente, Centre for Telematics and Information Technology (CTIT)

Fingerprint

Face recognition
Closed circuit television systems
Crime
Lighting

Keywords

  • METIS-286374
  • IR-80521
  • Face Recognition
  • SCS-Safety
  • EWI-21891
  • Performance Evaluation
  • Image quality

Cite this

Dutta, A., Veldhuis, R. N. J., & Spreeuwers, L. J. (2012). The Impact of Image Quality on the Performance of Face Recognition. In 33rd WIC Symposium on Information Theory in the Benelux (pp. 141-148). Enschede, the Netherlands: Centre for Telematics and Information Technology (CTIT).

Dutta, A.; Veldhuis, Raymond N.J.; Spreeuwers, Lieuwe Jan / The Impact of Image Quality on the Performance of Face Recognition.

33rd WIC Symposium on Information Theory in the Benelux. Enschede, the Netherlands : Centre for Telematics and Information Technology (CTIT), 2012. p. 141-148.

Research output: Scientific - peer-reviewConference contribution

@inbook{22eceaab097342738a047558763bcf68,
title = "The Impact of Image Quality on the Performance of Face Recognition",
abstract = "The performance of a face recognition system depends on the quality of both test and reference images participating in the face comparison process. In a forensic evaluation case involving face recognition, we do not have any control over the quality of the trace (image captured by a CCTV at a crime scene) and it is usually of very low quality. However, forensic investigators have some control over the quality of suspects reference images. In this paper, we investigate if it is useful to modify the quality of the reference (usually, good quality mug shot) images in order to achieve better recognition performance using a view based face recognition system. We found that approximately matching non-frontal pose between test and reference images can greatly improve recognition performance. Moreover, it is the relative difference in pose between the test and reference image that determines the extent of influence that other quality parameters like illumination, noise, motion blur, and resolution have on the face recognition performance.",
keywords = "METIS-286374, IR-80521, Face Recognition, SCS-Safety, EWI-21891, Performance Evaluation, Image quality",
author = "A. Dutta and Veldhuis, {Raymond N.J.} and Spreeuwers, {Lieuwe Jan}",
year = "2012",
month = "5",
isbn = "978-90-365-3383-6",
publisher = "Centre for Telematics and Information Technology (CTIT)",
pages = "141--148",
booktitle = "33rd WIC Symposium on Information Theory in the Benelux",
address = "Netherlands",

}

Dutta, A, Veldhuis, RNJ & Spreeuwers, LJ 2012, The Impact of Image Quality on the Performance of Face Recognition. in 33rd WIC Symposium on Information Theory in the Benelux. Centre for Telematics and Information Technology (CTIT), Enschede, the Netherlands, pp. 141-148.

The Impact of Image Quality on the Performance of Face Recognition. / Dutta, A.; Veldhuis, Raymond N.J.; Spreeuwers, Lieuwe Jan.

33rd WIC Symposium on Information Theory in the Benelux. Enschede, the Netherlands : Centre for Telematics and Information Technology (CTIT), 2012. p. 141-148.

Research output: Scientific - peer-reviewConference contribution

TY - CHAP

T1 - The Impact of Image Quality on the Performance of Face Recognition

AU - Dutta,A.

AU - Veldhuis,Raymond N.J.

AU - Spreeuwers,Lieuwe Jan

PY - 2012/5

Y1 - 2012/5

N2 - The performance of a face recognition system depends on the quality of both test and reference images participating in the face comparison process. In a forensic evaluation case involving face recognition, we do not have any control over the quality of the trace (image captured by a CCTV at a crime scene) and it is usually of very low quality. However, forensic investigators have some control over the quality of suspects reference images. In this paper, we investigate if it is useful to modify the quality of the reference (usually, good quality mug shot) images in order to achieve better recognition performance using a view based face recognition system. We found that approximately matching non-frontal pose between test and reference images can greatly improve recognition performance. Moreover, it is the relative difference in pose between the test and reference image that determines the extent of influence that other quality parameters like illumination, noise, motion blur, and resolution have on the face recognition performance.

AB - The performance of a face recognition system depends on the quality of both test and reference images participating in the face comparison process. In a forensic evaluation case involving face recognition, we do not have any control over the quality of the trace (image captured by a CCTV at a crime scene) and it is usually of very low quality. However, forensic investigators have some control over the quality of suspects reference images. In this paper, we investigate if it is useful to modify the quality of the reference (usually, good quality mug shot) images in order to achieve better recognition performance using a view based face recognition system. We found that approximately matching non-frontal pose between test and reference images can greatly improve recognition performance. Moreover, it is the relative difference in pose between the test and reference image that determines the extent of influence that other quality parameters like illumination, noise, motion blur, and resolution have on the face recognition performance.

KW - METIS-286374

KW - IR-80521

KW - Face Recognition

KW - SCS-Safety

KW - EWI-21891

KW - Performance Evaluation

KW - Image quality

M3 - Conference contribution

SN - 978-90-365-3383-6

SP - 141

EP - 148

BT - 33rd WIC Symposium on Information Theory in the Benelux

PB - Centre for Telematics and Information Technology (CTIT)

ER -

Dutta A, Veldhuis RNJ, Spreeuwers LJ. The Impact of Image Quality on the Performance of Face Recognition. In 33rd WIC Symposium on Information Theory in the Benelux. Enschede, the Netherlands: Centre for Telematics and Information Technology (CTIT). 2012. p. 141-148.