RMIT University
Browse

Newborn and infant discrimination: revisiting footprints

journal contribution
posted on 2024-11-02, 07:45 authored by Johannes Kotzerke, Stephen DavisStephen Davis, R. Hayes, Kathryn HoradamKathryn Horadam
Formal registration of newborns and infants is a necessity to secure their rights for health care or education and to support law enforcement agencies in the fight against the trafficking of children or their illegal adoption. Ideally, all these requirements can be met using a newborn and infant biometric. Difficulties arise due to the small size and fragility of the infants' physical structures and the effects of rapid physical growth. We review the literature for suitable biometrics and investigate the footprint; asking (a) if there is a time frame shortly after birth when the friction ridge skin pattern of a newborn can be reliably captured; and (b) if the footprint crease pattern is a suitable newborn and infant biometric. For (a), we were unable to confirm the existence of such a time frame. For (b), we performed automatic verification experiments on a small test set of 20 pairs of crease patterns, and then a larger test set, achieving EERs of 22.22% and 46.39%, respectively. The comparison of two dizygotic twins did not show any noteworthy performance difference. Based on these results we do not recommend the foot crease pattern as a newborn or infant biometric for automatic verification.

Funding

Novel dissimilarity techniques for characterising noisy spatial networks

Australian Research Council

Find out more...

History

Journal

Australian Journal of Forensic Sciences

Volume

51

Issue

1

Start page

95

End page

108

Total pages

14

Publisher

Taylor and Francis

Place published

United Kingdom

Language

English

Copyright

© 2017 Australian Academy of Forensic Sciences

Former Identifier

2006084230

Esploro creation date

2020-06-22

Fedora creation date

2019-01-31

Usage metrics

    Scholarly Works

    Exports

    RefWorks
    BibTeX
    Ref. manager
    Endnote
    DataCite
    NLM
    DC