Towards the automation of forensic facial individualisation: Comparing forensic to non forensic eyebrow features

Chris Zeinstra; Raymond Veldhuis; Luuk Spreeuwers

Towards the automation of forensic facial individualisation: Comparing forensic to non forensic eyebrow features

Chris Zeinstra, Raymond Veldhuis, Luuk Spreeuwers

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution › Academic

19 Downloads (Pure)

Abstract

The Facial Identification Scientific Working Group (FISWG) publishes recommendations regarding one-to-one facial comparisons. At this moment a draft version of a facial image comparison feature list for morphological analysis has been published. This feature list is based on casework experience by forensic facial examiners. This paper investigates whether the performance of the FISWG eyebrow feature set can be considered as being "state-of-the-art". We compare the recognition performance of one particular state-of-the-art non forensic eyebrow feature set to a semi-automated version of the forensic FISWG eyebrow feature set. The recognition performance is measured in terms of the forensic relevant log-likelihood-ratio cost metric $C_{llr}$. It is shown the FISWG feature set can be considered as being "state-of-the-art" and there actually exists a collection of feature sets that have similar performance.

Original language	English
Title of host publication	35rd WIC Symposium on Information Theory in the Benelux and The 4th WIC/IEEE Symposium on Information Theory and Signal Processing in the Benelux
Subtitle of host publication	Eindhoven, The Netherlands, May 12-13, 2014
Editors	Boris Skoric, Tanya Ignatenko
Place of Publication	Eindhoven
Publisher	Werkgemeenschap voor Informatie- en Communicatietheorie (WIC)
Pages	73-80
Number of pages	8
ISBN (Print)	978-90-365-3383-6
Publication status	Published - 12 May 2014
Event	35th WIC Symposium on Information Theory in the Benelux 2014 - Eindhoven, Netherlands Duration: 12 May 2014 → 13 May 2014 Conference number: 35

Conference

Conference	35th WIC Symposium on Information Theory in the Benelux 2014
Country	Netherlands
City	Eindhoven
Period	12/05/14 → 13/05/14

Keywords

SCS-Safety
METIS-305909
IR-91374
EWI-24819

Access to Document

Zeinstra2014towards
open
Final published version, 507 KB

Fingerprint Dive into the research topics of 'Towards the automation of forensic facial individualisation: Comparing forensic to non forensic eyebrow features'. Together they form a unique fingerprint.

Cite this

Zeinstra, C., Veldhuis, R., & Spreeuwers, L. (2014). Towards the automation of forensic facial individualisation: Comparing forensic to non forensic eyebrow features. In B. Skoric, & T. Ignatenko (Eds.), 35rd WIC Symposium on Information Theory in the Benelux and The 4th WIC/IEEE Symposium on Information Theory and Signal Processing in the Benelux: Eindhoven, The Netherlands, May 12-13, 2014 (pp. 73-80). Werkgemeenschap voor Informatie- en Communicatietheorie (WIC).

Zeinstra, Chris ; Veldhuis, Raymond ; Spreeuwers, Luuk. / Towards the automation of forensic facial individualisation : Comparing forensic to non forensic eyebrow features. 35rd WIC Symposium on Information Theory in the Benelux and The 4th WIC/IEEE Symposium on Information Theory and Signal Processing in the Benelux: Eindhoven, The Netherlands, May 12-13, 2014. editor / Boris Skoric ; Tanya Ignatenko. Eindhoven : Werkgemeenschap voor Informatie- en Communicatietheorie (WIC), 2014. pp. 73-80

@inproceedings{72703a0d092f440f9fd8e6474a0d8d5a,

title = "Towards the automation of forensic facial individualisation: Comparing forensic to non forensic eyebrow features",

abstract = "The Facial Identification Scientific Working Group (FISWG) publishes recommendations regarding one-to-one facial comparisons. At this moment a draft version of a facial image comparison feature list for morphological analysis has been published. This feature list is based on casework experience by forensic facial examiners. This paper investigates whether the performance of the FISWG eyebrow feature set can be considered as being {"}state-of-the-art{"}. We compare the recognition performance of one particular state-of-the-art non forensic eyebrow feature set to a semi-automated version of the forensic FISWG eyebrow feature set. The recognition performance is measured in terms of the forensic relevant log-likelihood-ratio cost metric $C_{llr}$. It is shown the FISWG feature set can be considered as being {"}state-of-the-art{"} and there actually exists a collection of feature sets that have similar performance.",

keywords = "SCS-Safety, METIS-305909, IR-91374, EWI-24819",

author = "Chris Zeinstra and Raymond Veldhuis and Luuk Spreeuwers",

note = "http://www.win.tue.nl/sitb2014/proceedings.pdf; 35th WIC Symposium on Information Theory in the Benelux 2014 ; Conference date: 12-05-2014 Through 13-05-2014",

year = "2014",

month = may,

day = "12",

language = "English",

isbn = "978-90-365-3383-6",

pages = "73--80",

editor = "Boris Skoric and Tanya Ignatenko",

booktitle = "35rd WIC Symposium on Information Theory in the Benelux and The 4th WIC/IEEE Symposium on Information Theory and Signal Processing in the Benelux",

publisher = "Werkgemeenschap voor Informatie- en Communicatietheorie (WIC)",

address = "Netherlands",

}

Zeinstra, C , Veldhuis, R & Spreeuwers, L 2014, Towards the automation of forensic facial individualisation: Comparing forensic to non forensic eyebrow features. in B Skoric & T Ignatenko (eds), 35rd WIC Symposium on Information Theory in the Benelux and The 4th WIC/IEEE Symposium on Information Theory and Signal Processing in the Benelux: Eindhoven, The Netherlands, May 12-13, 2014. Werkgemeenschap voor Informatie- en Communicatietheorie (WIC), Eindhoven, pp. 73-80, 35th WIC Symposium on Information Theory in the Benelux 2014, Eindhoven, Netherlands, 12/05/14.

Towards the automation of forensic facial individualisation : Comparing forensic to non forensic eyebrow features. / Zeinstra, Chris ; Veldhuis, Raymond ; Spreeuwers, Luuk.

35rd WIC Symposium on Information Theory in the Benelux and The 4th WIC/IEEE Symposium on Information Theory and Signal Processing in the Benelux: Eindhoven, The Netherlands, May 12-13, 2014. ed. / Boris Skoric; Tanya Ignatenko. Eindhoven : Werkgemeenschap voor Informatie- en Communicatietheorie (WIC), 2014. p. 73-80.

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution › Academic

TY - GEN

T1 - Towards the automation of forensic facial individualisation

T2 - 35th WIC Symposium on Information Theory in the Benelux 2014

AU - Zeinstra, Chris

AU - Veldhuis, Raymond

AU - Spreeuwers, Luuk

N1 - Conference code: 35

PY - 2014/5/12

Y1 - 2014/5/12

N2 - The Facial Identification Scientific Working Group (FISWG) publishes recommendations regarding one-to-one facial comparisons. At this moment a draft version of a facial image comparison feature list for morphological analysis has been published. This feature list is based on casework experience by forensic facial examiners. This paper investigates whether the performance of the FISWG eyebrow feature set can be considered as being "state-of-the-art". We compare the recognition performance of one particular state-of-the-art non forensic eyebrow feature set to a semi-automated version of the forensic FISWG eyebrow feature set. The recognition performance is measured in terms of the forensic relevant log-likelihood-ratio cost metric $C_{llr}$. It is shown the FISWG feature set can be considered as being "state-of-the-art" and there actually exists a collection of feature sets that have similar performance.

AB - The Facial Identification Scientific Working Group (FISWG) publishes recommendations regarding one-to-one facial comparisons. At this moment a draft version of a facial image comparison feature list for morphological analysis has been published. This feature list is based on casework experience by forensic facial examiners. This paper investigates whether the performance of the FISWG eyebrow feature set can be considered as being "state-of-the-art". We compare the recognition performance of one particular state-of-the-art non forensic eyebrow feature set to a semi-automated version of the forensic FISWG eyebrow feature set. The recognition performance is measured in terms of the forensic relevant log-likelihood-ratio cost metric $C_{llr}$. It is shown the FISWG feature set can be considered as being "state-of-the-art" and there actually exists a collection of feature sets that have similar performance.

KW - SCS-Safety

KW - METIS-305909

KW - IR-91374

KW - EWI-24819

M3 - Conference contribution

SN - 978-90-365-3383-6

SP - 73

EP - 80

BT - 35rd WIC Symposium on Information Theory in the Benelux and The 4th WIC/IEEE Symposium on Information Theory and Signal Processing in the Benelux

A2 - Skoric, Boris

A2 - Ignatenko, Tanya

PB - Werkgemeenschap voor Informatie- en Communicatietheorie (WIC)

CY - Eindhoven

Y2 - 12 May 2014 through 13 May 2014

ER -

Zeinstra C , Veldhuis R , Spreeuwers L. Towards the automation of forensic facial individualisation: Comparing forensic to non forensic eyebrow features. In Skoric B, Ignatenko T, editors, 35rd WIC Symposium on Information Theory in the Benelux and The 4th WIC/IEEE Symposium on Information Theory and Signal Processing in the Benelux: Eindhoven, The Netherlands, May 12-13, 2014. Eindhoven: Werkgemeenschap voor Informatie- en Communicatietheorie (WIC). 2014. p. 73-80