Cosmetic applied based face recognition for biometric passport

Ziaul Haque Choudhury (Department of Information Technology, B.S. Abdur Rahman Crescent Institute of Science and Technology, Chennai, India)
M. Munir Ahamed Rabbani (Department of Mathematics, B.S. Abdur Rahman Crescent Institute of Science and Technology, Chennai, India)

International Journal of Intelligent Unmanned Systems

ISSN: 2049-6427

Publication date: 29 July 2019

Abstract

Purpose

Nowadays, the use of forged e-passport is increasing, which is threatening national security. It is important to improve the national security against international crime or terrorism. There is a weak verification process caused by lack of identification processes such as a physical check, biometric check and electronic check. The e-passport can prevent the passport cloning or forging resulting from the illegal immigration. The paper aims to discuss these issues.

Design/methodology/approach

This paper focuses on face recognition to improve the biometric authentication for an e-passport, and it also introduces facial permanent mark detection from the makeup or cosmetic-applied faces, twins and similar faces. An algorithm is proposed to detect the cosmetic-applied facial permanent marks such as mole, freckle, birthmark and pockmark. Active Shape Model into Active Appearance Model using Principal Component Analysis is applied to detect the facial landmarks. Facial permanent marks are detected by applying the Canny edge detector and Gradient Field Histogram of Oriented Gradient.

Findings

This paper demonstrated an algorithm and proposed facial marks detection from cosmetic or makeup-applied faces for a secure biometric passport in the field of personal identification for national security. It also presented to detect and identify identical twins and similar faces. This paper presented facial marks detection from the cosmetic-applied face, which can be mixed with traditional methods. However, the use of the proposed technique faced some challenges due to the use of cosmetic. The combinations of the algorithm for facial mark recognition matching with classical methods were able to attain lower errors in this proposed experiment.

Originality/value

The proposed method will enhance the national security and it will improve the biometric authentication for the e-passport. The proposed algorithm is capable of identifying facial marks from cosmetic-applied faces accurately, with less false positives. The proposed technique shows the best results.

Keywords

Citation

Choudhury, Z.H. and Rabbani, M.M.A. (2019), "Cosmetic applied based face recognition for biometric passport", International Journal of Intelligent Unmanned Systems, Vol. 8 No. 1, pp. 3-22. https://doi.org/10.1108/IJIUS-02-2019-0016

Publisher

:

Emerald Publishing Limited

Copyright © 2019, Emerald Publishing Limited

To read the full version of this content please select one of the options below

You may be able to access this content by logging in via Shibboleth, Open Athens or with your Emerald account.
To rent this content from Deepdyve, please click the button.
If you think you should have access to this content, click the button to contact our support team.