Abstract
Between the ever-increasing availability of surveillance evidence and expert-based forensic facial comparison being considered admissible in court, confirming its validity is paramount. Facial comparison is most commonly conducted using morphological analysis (MA), a largely untested feature-based approach. This study aimed at validating the current recommended practice of MA in both standardised and suboptimal surveillance samples. Face pools of 175 South African males were compiled with a series of facial photographs, using images from the Wits Face Database. The first 75 face pools consisted of wildtype (unstandardised) high-quality target photographs, while the remaining 100 face pools consisted of suboptimal closed-circuit television (CCTV) target images. Target images were compared to high-quality standardised photographs. Face pools were analysed using the Facial Identification Scientific Working Group’s guidelines and feature list. Confusion matrices were used to determine the performance of MA in each cohort. MA was found highly accurate (chance-corrected accuracy (CCA): 99.1%) and reliable (κ = 0.921) in the photographic sample and less accurate (CCA: 82.6%) and reliable (κ = 0.743), in the CCTV sample. Higher false-positive and false-negative rates were noted for the CCTV sample, with the majority of errors resulting in false-negative outcomes. The decreased performance in the CCTV sample was attributed to various factors including image quality, angle of recording and lighting. Other studies testing facial comparison identified lower accuracies and reliability across various conditions. Better performance was found here and in other studies that included some form of facial feature list, reinforcing the importance of using a systematic facial feature list.
Similar content being viewed by others
Availability of data and material
The datasets analysed during the current study are not publicly available due the sensitive nature of the Wits Face Database. The database has the identifier https://doi.org/10.17605/OSF.IO/Q8V2R. A sample of the database (consisting of one of the author’s’ facial images and recordings) is made available at https://hdl.handle.net/10539/29924. Restrictions apply to accessing the database or part thereof, which was used under license for the current study and, hence, it is not publicly available. Access is limited to formal request and conditional approval by the School of Anatomical Sciences Collections Committee, strictly for non-commercial research.
References
Davis JP, Valentine T, Wilkinson C (2012) Facial image comparison. In: Wilkinson C, Rynn C (eds) Craniofacial identification, 1st edn. Cambridge University Press, New York, pp 136–153
Lee W-L, Wilkinson C, Memon A, Houston K (2009) Matching unfamiliar faces from poor quality closed-circuit television (CCTV) footage: an evaluation of the effect of training on facial identification ability. AXIS Online J Cent Anat Hum Identif 1:19–28
Jain AK, Klare B, Park U (2012) Face matching and retrieval in forensics applications. IEEE Multimed 19:20–27. https://doi.org/10.1109/MMUL.2012.4
Phillips PJ (2011) Improving face recognition technology. Computer (Long Beach Calif) 44:84–86. https://doi.org/10.1109/MC.2011.87
Urbanová P (2016) Performance of distance-based matching algorithms in 3D facial identification. Egypt J Forensic Sci 6:135–151. https://doi.org/10.1016/j.ejfs.2016.04.004
Davis JP, Valentine T, Davis RE (2010) Computer assisted photo-anthropometric analyses of full-face and profile facial images. Forensic Sci Int 200:165–176. https://doi.org/10.1016/j.forsciint.2010.04.012
Heyer R, Semmler C (2013) Forensic confirmation bias: the case of facial image comparison. J Appl Res Mem Cogn 2:68–70. https://doi.org/10.1016/j.jarmac.2013.01.008
Caple J, Stephan CN (2016) A standardized nomenclature for craniofacial and facial anthropometry. Int J Legal Med 130:863–879. https://doi.org/10.1007/s00414-015-1292-1
Kleinberg KF, Vanezis P, Burton AM (2007) Failure of anthropometry as a facial identification technique using high-quality photographs. J Forensic Sci 52:779–783. https://doi.org/10.1111/j.1556-4029.2007.00458.x
Moreton R, Morley J (2011) Investigation into the use of photoanthropometry in facial image comparison. Forensic Sci Int 212:231–237. https://doi.org/10.1016/j.forsciint.2011.06.023
Stephan CN, Caple JM, Guyomarc’h P, Claes P (2019) An overview of the latest developments in facial imaging. Forensic Sci Res 4:10–28. https://doi.org/10.1080/20961790.2018.1519892
Facial Identification Scientific Working Group (2019) Facial comparison overview and methodology guidelines. Available at: https://fiswg.org/fiswg_facial_comparison_overview_and_methodology_guidelines_V1.0_20191025.pdf
Houlton TMR, Steyn M (2018) Finding Makhubu: a morphological forensic facial comparison. Forensic Sci Int 285:13–20. https://doi.org/10.1016/J.FORSCIINT.2018.01.022
Steyn M, Pretorius M, Briers N et al (2018) Forensic facial comparison in South Africa: state of the science. Forensic Sci Int 287:190–194. https://doi.org/10.1016/J.FORSCIINT.2018.04.006
Burton AM, Wilson S, Cowan M, Bruce V (1999) Research article evidence from security surveillance. Psychol Sci 10:243–248. https://doi.org/10.1111/1467-9280.00144
Bruce V, Henderson Z, Newman C, Burton AM (2001) Matching identities of familiar and unfamiliar faces caught on CCTV images. J Exp Psychol Appl 7:207–218. https://doi.org/10.1037//1076-898X.7.3.207
Burton AM, Wilson S, Cowan M, Bruce V (1999) Face recognition in poor-quality video: evidence from security surveillance. Psychol Sci 10:243–248. https://doi.org/10.1111/1467-9280.00144
Henderson Z, Bruce V, Burton AM (2001) Matching the faces of robbers captured on video. Appl Cogn Psychol 15:445–464. https://doi.org/10.1002/acp.718
Megreya AM, Burton AM (2008) Matching faces to photographs: poor performance in eyewitness memory (without the memory). J Exp Psychol Appl 14:364–372. https://doi.org/10.1037/a0013464
Megreya AM, Memon A, Havard C (2012) The headscarf effect: direct evidence from the eyewitness identification paradigm. Appl Cogn Psychol 26:308–315. https://doi.org/10.1002/acp.1826
Wilkinson C, Evans R (2009) Are facial image analysis experts any better than the general public at identifying individuals from CCTV images? Sci Justice 49:191–196. https://doi.org/10.1016/j.scijus.2008.10.011
Burton M, White D, McNeill A (2010) The Glasgow face matching test. Behav Res Methods 42:286–291. https://doi.org/10.3758/BRM.42.1.286
Megreya AM, Bindemann M, Havard C (2011) Sex differences in unfamiliar face identification: evidence from matching tasks. Acta Psychol (Amst) 137:83–89. https://doi.org/10.1016/j.actpsy.2011.03.003
Ritz-Timme S, Gabriel P, Obertovà Z et al (2011) A new atlas for the evaluation of facial features: advantages, limits, and applicability. Int J Legal Med 125:301–306. https://doi.org/10.1007/s00414-010-0446-4
Ritz-Timme S, Gabriel P, Tutkuviene J et al (2011) Metric and morphological assessment of facial features: a study on three European populations. Forensic Sci Int 207:239.e1-239.e8. https://doi.org/10.1016/j.forsciint.2011.01.035
Towler A, White D, Kemp RI (2017) Evaluating the feature comparison strategy for forensic face identification. J Exp Psychol Appl 23:47–58. https://doi.org/10.1037/xap0000108
Megreya AM, Bindemann M (2018) Feature instructions improve face-matching accuracy. PLoS One 13:1–16. https://doi.org/10.1371/journal.pone.0193455
Porter G, Doran G (2000) An anatomical and photographic technique for forensic facial identification. Forensic Sci Int 114:97–105. https://doi.org/10.1016/S0379-0738(00)00290-5
Facial Identification Scientific Working Group (2018) Facial image comparison feature list for morphological analysis. Available at: https://fiswg.org/FISWG_Morph_Analysis_Feature_List_v2.0_20180911.pdf
Bacci N, Davimes J, Steyn M, Briers N (2020) Wits face database. Wits Institutional Repos. Environ. Dsp, In https://hdl.handle.net/10539/29924
Steyn M, Smith JR (2007) Interpretation of ante-mortem stature estimates in South Africans. Forensic Sci Int 171:97–102. https://doi.org/10.1016/j.forsciint.2006.10.006
Speckeis C (2011) Can ACE-V be validated? J Forensic Identif 61:201–209 Available at: https://www.ncjrs.gov/pdffiles1/nij/225329.pdf
R Core Team (2020) R: a language and environment for statistical computing. Available at: https://www.r-project.org/
Watson PF, Petrie A (2010) Method agreement analysis: a review of correct methodology. Theriogenol 73:1167–1179. https://doi.org/10.1016/j.theriogenology.2010.01.003
Kuhn M (2008) Building predictive models in R using the caret package. J Stat Softw 28:1–26. https://doi.org/10.18637/jss.v028.i05
Kuhn M (2020) Caret: classification and regression training R package. Available at: https://cran.r-project.org/package=caret
Dietterich TG (1998) Approximate statistical tests for comparing supervised classification learning algorithms. Neural Comput 10:1895–1923. https://doi.org/10.1007/978-3-319-50926-6_6
Norris C, McCahill M, Wood D (2004) Editorial. The growth of CCTV: a global perspective on the international diffusion of video surveillance in publicly accessible space. Surveill Soc 2:110–135. https://doi.org/10.24908/ss.v2i2/3.3369
Edmond G (2013) Just truth? Carefully applying history, philosophy and sociology of science to the forensic use of CCTV images. Stud Hist Philos Sci Part C Stud Hist Philos Biol Biomed Sci 44:80–91. https://doi.org/10.1016/j.shpsc.2012.09.004
Bromby M (2003) At face value? New Law J Expert Witn Suppl 28:302–303
Dewhurst SA, Hay DC, Wickham LHV (2005) Distinctiveness, typicality, and recollective experience in face recognition: a principal components analysis. Psychon Bull Rev 12:1032–1037. https://doi.org/10.3758/BF03206439
Lee K, Byatt G, Rhodes G (2000) Caricature effects, distinctiveness, and identification: testing the face-space framework. Psychol Sci 11:379–385. https://doi.org/10.1111/1467-9280.00274
Thompson WC, Taroni F, Aitken CGG (2003) How the probability of a false positive affects the value of DNA evidence. J Forensic Sci 48:2001171. https://doi.org/10.1520/jfs2001171
Gibelli D, Obertová Z, Ritz-Timme S et al (2016) The identification of living persons on images: a literature review. Leg Med 19:52–60. https://doi.org/10.1016/j.legalmed.2016.02.001
Damjanovski V (2014) CCTV from light to pixels, 3rd edn. Elsevier, Oxford
Ward D (2013) Testing camera height vs image quality. Pennsylvania, USA Available at: https://ipvm.com/reports/testing-camera-height
Cohen N, Gattuso J, MacLennan-Brown K (2009) CCTV Operational requirements manual. Home Office Scientific Development Branch. Sandridge, UK Available at: https://assets.publishing.service.gov.uk/government/uploads/system/uploads/attachment_data/file/378443/28_09_CCTV_OR_Manual2835.pdf
Zhang X, Gao Y (2009) Face recognition across pose: a review. Pattern Recognit 42:2876–2896. https://doi.org/10.1016/j.patcog.2009.04.017
Burton A, Bruce V, Hancock PJB (1999) From pixels to people: a model of familiar face recognition. Cogn Sci 23:1–31. https://doi.org/10.1016/S0364-0213(99)80050-0
Bruce V (1994) Stability from variation: the case of face recognition The M.D. Vernon Memorial Lecture. Q J Exp Psychol Sect A 47:5–28. https://doi.org/10.1080/14640749408401141
Davis JP, Maigut A, Forrest C (2019) The wisdom of the crowd: a case of post- to ante-mortem face matching by police super-recognisers. Forensic Sci Int 302:109910. https://doi.org/10.1016/j.forsciint.2019.109910
Roelofse MM, Steyn M, Becker PJ (2008) Photo identification: facial metrical and morphological features in South African males. Forensic Sci Int 177:168–175. https://doi.org/10.1016/j.forsciint.2007.12.003
Vanezis P, Lu D, Cockburn J et al (1996) Morphological classification of facial features in adult Caucasian males based on an assessment of photographs of 50 subjects. J Forensic Sci 41:13998J. https://doi.org/10.1520/jfs13998j
Dodd V (2018) UK police use of facial recognition technology a failure, says report. Guard. Available at: https://www.theguardian.com/uk-news/2018/may/15/uk-police-use-of-facial-recognition-technology-failure
Press Association (2018) Welsh police wrongly identify thousands as potential criminals. Guard. Available at: https://www.theguardian.com/uk-news/2018/may/05/welsh-police-wrongly-identify-thousands-as-potential-criminals
Wall M (2019) Biased and wrong? Facial recognition tech in the dock. BBC News 1–12. Available at: https://www.bbc.com/news/business-48842750
Grother P, Ngan M, Hanaoka K (2019) Face recognition vendor test part 3: demographic effects. Gaithersburg, MD. https://doi.org/10.6028/NIST.IR.8280
Morrison R (2019) ‘Racist’ facial recognition technology used in law enforcement, banking and schools misidentifies African American and Asian people 100 times more often than whites, study shows. DailyMail Online. Available at: https://www.dailymail.co.uk/sciencetech/article-7811779/Massive-errors-facial-recognition-tech-US-study.html
Belhumeur PN, Jacobs DW, Kriegman DJ, Kumar N (2013) Localizing parts of faces using a consensus of exemplars. IEEE Trans Pattern Anal Mach Intell 35:2930–2940. https://doi.org/10.1109/TPAMI.2013.23
Burgos-Artizzu XP, Perona P, Dollar P (2013) Robust face landmark estimation under occlusion. Proc IEEE Int Conf Comput Vis:1513–1520. https://doi.org/10.1109/ICCV.2013.191
Sagonas C, Antonakos E, Tzimiropoulos G et al (2015) 300 faces In-The-Wild Challenge: database and results. Image Vis Comput 47:3–18. https://doi.org/10.1016/j.imavis.2016.01.002
Ellis HD, Shepherd JW, Davies GM (1979) Identification of familiar and unfamiliar faces from internal and external features: some implications for theories of face recognition. Perception 8:431–439. https://doi.org/10.1068/p080431
Megreya AM, Burton AM (2006) Unfamiliar faces are not faces evidence from a matching task. Mem Cognit 34:865–876. https://doi.org/10.3758/BF03193433
Bindemann M, Johnston RA (2017) Understanding how unfamiliar faces become familiar: Introduction to a special issue on face learning. Q J Exp Psychol 70:859–862. https://doi.org/10.1080/17470218.2016.1267235
Facial Identification Scientific Working Group (2012) FISWG guidelines for facial comparison methods. Available at: https://www.fiswg.org/FISWG_GuidelinesforFacialComparisonMethods_v1.0_2012_02_02.pdf
Alenezi HM, Bindemann M, Fysh MC (2015) Johnston RA (2015) Face matching in a long task: Enforced rest and desk-switching cannot maintain identification accuracy. PeerJ 3:e1184. https://doi.org/10.7717/peerj.1184
Acknowledgements
Thanks are due to all the participants who agreed to be photographed and recorded for the development of the Wits Face Database from which the face pool samples were composed. Special thanks also go to Joshua Davimes for his instrumental contributions in developing the Wits Face Database. We are also especially grateful towards Tamara Lottering for her assistance in composing the face pools and Gideon LeRoux for his assistance in the capturing and extracting of the CCTV recordings from the University security systems. Thanks are also due to the volunteers who aided in participant recruitment: Jesse Fredericks, Kiveshen Pillay, Rethabile Masiu, Sameerah Sallie, Daniel Munesamy, Laurette Joubert, Jordan Swiegers, Betty Mkabela, Johannes P. Meyer, Amy Spies, Natasha Loubser, Nicole Virgili, Dan-Joel Lukumbi, Tamara Lottering, Mathabatha Ntjie, Claudia Landsman, Raheema Dalika, Merete Goosen, Stephanie Souris, Rabelani Negota, Mahlatse Mahasha, and Jessica Manavhela.
Funding
The current study was conducted with support from the South African National Research Foundation (NRF) and the J.J.J. Smieszeck Fellowship from the School of Anatomical Sciences, University of the Witwatersrand (DAAD-NRF and J.J.J. Smieszeck Fellowship funds awarded to N. Bacci (Grant No.: 11858) and NRF funds awarded to N. Briers as part of the Improving Methodologies and Practices in Craniofacial Identification (Grant No.: CSUR160425163022; UID: 106031). Any opinions, findings, and conclusions or recommendations expressed in this study are those of the authors and therefore the NRF and University of the Witwatersrand, Johannesburg, do not accept any liability in regard thereto.
Author information
Authors and Affiliations
Contributions
Conceptualisation: Nicholas Bacci, Maryna Steyn, and Nanette Briers. Methodology: Nicholas Bacci, Maryna Steyn, Nanette Briers, and Tobias Houlton. Formal analysis and investigation: Nicholas Bacci and Tobias Houlton. Writing (original draft preparation): Nicholas Bacci. Writing (review and editing): Nicholas Bacci, Maryna Steyn, Nanette Briers, and Tobias Houlton. Funding acquisition: Nicholas Bacci, Maryna Steyn, and Nanette Briers. Resources: Maryna Steyn and Nanette Briers. Supervision: Maryna Steyn and Nanette Briers.
Corresponding author
Ethics declarations
Ethics approval and consent to participate
The study was unconditionally approved by the Human Research Ethics Committee (Medical) of the University of the Witwatersrand, Johannesburg. Ethics clearance certificate number: M171026. All participants included in the study signed an informed consent form granting the authors the use of their data.
Consent for publication
No participant information or photographs are published as part of the study, as agreed upon in the aforementioned informed consent signed for participation.
Conflict of interest
The authors declare no conflict of interest.
Code availability
Not applicable
Additional information
Publisher’s note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
Rights and permissions
About this article
Cite this article
Bacci, N., Houlton, T.M.R., Briers, N. et al. Validation of forensic facial comparison by morphological analysis in photographic and CCTV samples. Int J Legal Med 135, 1965–1981 (2021). https://doi.org/10.1007/s00414-021-02512-3
Received:
Accepted:
Published:
Issue Date:
DOI: https://doi.org/10.1007/s00414-021-02512-3