Computation of likelihood ratio from small sample set of within-source variability

Research output: Chapter in Book/Report/Conference proceedingConference contributionAcademic

19 Downloads (Pure)

Abstract

In this paper we describe a new method of likelihood ratio computation for score-based biometric recognition systems given a small number of samples in within-source variability dataset. Generally the number of samples in within-source variability dataset is less than the number of samples in between-source variability dataset and therefore the probability density function (pdf) of within-source variability dataset cannot be estimated reliably compared to the pdf of between-source variability. The proposed method estimates the pdf of within-source variability from estimates of the within-source variability mean and variance and the pdf of between-source variability by minimizing the Kullback-Leibler distance [1] of the pdf of the within-source variability to that of the between-source variability given within-source variability mean and variance. It thus finds a conservative estimate of the pdf of within-source variability. Working out this optimization problem results in an log likelihood ration that is a second order polynomial of a given score value. We apply this approach of likelihood ratio computation in the area of face recognition. An existing commercial face recognition system [2] is used to obtain scores for the sets of within-source variability and between-source variability from a set of image data taken from SCFace database [3]. It contains images taken in uncontrolled indoor environment using five video surveillance cameras of various qualities. For each subject, there are also mug shots taken in same conditions as would be expected for any law enforcement or national security use. We explore the feasibility of using an existing biometric face recognition system in forensic application by discussing some specific cases in forensic framework.
Original languageUndefined
Title of host publication6th European Academy of Forensic Science Conference (EAFS 2012)
Place of PublicationThe Hague
PublisherEuropean Academy of Forensic Science
Pages-
Number of pages1
ISBN (Print)not assigned
Publication statusPublished - Aug 2012

Publication series

Name
PublisherEuropean Academy of Forensic Science
NumberTowards Fo

Keywords

  • SCS-Safety
  • IR-83416
  • METIS-289789
  • EWI-22536

Cite this

Ali, T., Spreeuwers, L. J., & Veldhuis, R. N. J. (2012). Computation of likelihood ratio from small sample set of within-source variability. In 6th European Academy of Forensic Science Conference (EAFS 2012) (pp. -). The Hague: European Academy of Forensic Science.
Ali, Tauseef ; Spreeuwers, Lieuwe Jan ; Veldhuis, Raymond N.J. / Computation of likelihood ratio from small sample set of within-source variability. 6th European Academy of Forensic Science Conference (EAFS 2012). The Hague : European Academy of Forensic Science, 2012. pp. -
@inproceedings{432487e92f43458585d080adaa605bd6,
title = "Computation of likelihood ratio from small sample set of within-source variability",
abstract = "In this paper we describe a new method of likelihood ratio computation for score-based biometric recognition systems given a small number of samples in within-source variability dataset. Generally the number of samples in within-source variability dataset is less than the number of samples in between-source variability dataset and therefore the probability density function (pdf) of within-source variability dataset cannot be estimated reliably compared to the pdf of between-source variability. The proposed method estimates the pdf of within-source variability from estimates of the within-source variability mean and variance and the pdf of between-source variability by minimizing the Kullback-Leibler distance [1] of the pdf of the within-source variability to that of the between-source variability given within-source variability mean and variance. It thus finds a conservative estimate of the pdf of within-source variability. Working out this optimization problem results in an log likelihood ration that is a second order polynomial of a given score value. We apply this approach of likelihood ratio computation in the area of face recognition. An existing commercial face recognition system [2] is used to obtain scores for the sets of within-source variability and between-source variability from a set of image data taken from SCFace database [3]. It contains images taken in uncontrolled indoor environment using five video surveillance cameras of various qualities. For each subject, there are also mug shots taken in same conditions as would be expected for any law enforcement or national security use. We explore the feasibility of using an existing biometric face recognition system in forensic application by discussing some specific cases in forensic framework.",
keywords = "SCS-Safety, IR-83416, METIS-289789, EWI-22536",
author = "Tauseef Ali and Spreeuwers, {Lieuwe Jan} and Veldhuis, {Raymond N.J.}",
note = "http://www.eafs2012.eu/home.html",
year = "2012",
month = "8",
language = "Undefined",
isbn = "not assigned",
publisher = "European Academy of Forensic Science",
number = "Towards Fo",
pages = "--",
booktitle = "6th European Academy of Forensic Science Conference (EAFS 2012)",

}

Ali, T, Spreeuwers, LJ & Veldhuis, RNJ 2012, Computation of likelihood ratio from small sample set of within-source variability. in 6th European Academy of Forensic Science Conference (EAFS 2012). European Academy of Forensic Science, The Hague, pp. -.

Computation of likelihood ratio from small sample set of within-source variability. / Ali, Tauseef; Spreeuwers, Lieuwe Jan; Veldhuis, Raymond N.J.

6th European Academy of Forensic Science Conference (EAFS 2012). The Hague : European Academy of Forensic Science, 2012. p. -.

Research output: Chapter in Book/Report/Conference proceedingConference contributionAcademic

TY - GEN

T1 - Computation of likelihood ratio from small sample set of within-source variability

AU - Ali, Tauseef

AU - Spreeuwers, Lieuwe Jan

AU - Veldhuis, Raymond N.J.

N1 - http://www.eafs2012.eu/home.html

PY - 2012/8

Y1 - 2012/8

N2 - In this paper we describe a new method of likelihood ratio computation for score-based biometric recognition systems given a small number of samples in within-source variability dataset. Generally the number of samples in within-source variability dataset is less than the number of samples in between-source variability dataset and therefore the probability density function (pdf) of within-source variability dataset cannot be estimated reliably compared to the pdf of between-source variability. The proposed method estimates the pdf of within-source variability from estimates of the within-source variability mean and variance and the pdf of between-source variability by minimizing the Kullback-Leibler distance [1] of the pdf of the within-source variability to that of the between-source variability given within-source variability mean and variance. It thus finds a conservative estimate of the pdf of within-source variability. Working out this optimization problem results in an log likelihood ration that is a second order polynomial of a given score value. We apply this approach of likelihood ratio computation in the area of face recognition. An existing commercial face recognition system [2] is used to obtain scores for the sets of within-source variability and between-source variability from a set of image data taken from SCFace database [3]. It contains images taken in uncontrolled indoor environment using five video surveillance cameras of various qualities. For each subject, there are also mug shots taken in same conditions as would be expected for any law enforcement or national security use. We explore the feasibility of using an existing biometric face recognition system in forensic application by discussing some specific cases in forensic framework.

AB - In this paper we describe a new method of likelihood ratio computation for score-based biometric recognition systems given a small number of samples in within-source variability dataset. Generally the number of samples in within-source variability dataset is less than the number of samples in between-source variability dataset and therefore the probability density function (pdf) of within-source variability dataset cannot be estimated reliably compared to the pdf of between-source variability. The proposed method estimates the pdf of within-source variability from estimates of the within-source variability mean and variance and the pdf of between-source variability by minimizing the Kullback-Leibler distance [1] of the pdf of the within-source variability to that of the between-source variability given within-source variability mean and variance. It thus finds a conservative estimate of the pdf of within-source variability. Working out this optimization problem results in an log likelihood ration that is a second order polynomial of a given score value. We apply this approach of likelihood ratio computation in the area of face recognition. An existing commercial face recognition system [2] is used to obtain scores for the sets of within-source variability and between-source variability from a set of image data taken from SCFace database [3]. It contains images taken in uncontrolled indoor environment using five video surveillance cameras of various qualities. For each subject, there are also mug shots taken in same conditions as would be expected for any law enforcement or national security use. We explore the feasibility of using an existing biometric face recognition system in forensic application by discussing some specific cases in forensic framework.

KW - SCS-Safety

KW - IR-83416

KW - METIS-289789

KW - EWI-22536

M3 - Conference contribution

SN - not assigned

SP - -

BT - 6th European Academy of Forensic Science Conference (EAFS 2012)

PB - European Academy of Forensic Science

CY - The Hague

ER -

Ali T, Spreeuwers LJ, Veldhuis RNJ. Computation of likelihood ratio from small sample set of within-source variability. In 6th European Academy of Forensic Science Conference (EAFS 2012). The Hague: European Academy of Forensic Science. 2012. p. -