View Complete Reference

Iorliam, A and Shangbum, FC (2017)

On the Use of Benfordís Law to Detect JPEG Biometric Data Tampering

Journal of Information Security 8, pp. 240-256.

ISSN/ISBN: Not available at this time. DOI: 10.4236/jis.2017.83016



Abstract: Tampering of biometric data has attracted a great deal of attention recently. Furthermore, there could be an intentional or accidental use of a particular biometric sample instead of another for a particular application. Therefore, there exists a need to propose a method to detect data tampering, as well as differentiate biometric samples in cases of intentional or accidental use for a different application. In this paper, fingerprint image tampering is studied. Furthermore, optically acquired fingerprints, synthetically generated fingerprints and contact-less acquired fingerprints are studied for separation purposes using the Benfordís law divergence metric. Benfordís law has shown in literature to be very effective in detecting tampering of natural images. In this paper, the Benfordís law features with support vector machine are proposed for the detection of malicious tampering of JPEG fingerprint images. This method is aimed at protecting against insider attackers and hackers. This proposed method detected tampering effectively, with Equal Error Rate (EER) of 2.08%. Again, the experimental results illustrate that, optically acquired fingerprints, synthetically generated fingerprints and contact-less acquired fingerprints can be separated by the proposed method effectively.


Bibtex:
@article {, AUTHOR = {Aamo Iorliam and F. Caleb Shangbum}, TITLE = {On the Use of Benfordís Law to Detect JPEG Biometric Data Tampering}, JOURNAL = {Journal of Information Security}, YEAR = {2017}, VOLUME = {8}, NUMBER = {}, PAGES = {240--256}, DOI = {10.4236/jis.2017.83016}, URL = {https://www.scirp.org/journal/PaperInformation.aspx?PaperID=77757}, }


Reference Type: Journal Article

Subject Area(s): Image Processing