Elsevier

Neurocomputing

Volume 100, 16 January 2013, Pages 153-162
Neurocomputing

Consent biometrics

https://doi.org/10.1016/j.neucom.2011.12.044Get rights and content

Abstract

Biometrics identifies/verifies a person using his/her physiological or behavioral characteristics. It is becoming an important ally for law enforcement and homeland security. However, there are some safety and privacy concerns: biometrics guarded system can be accessed when users are under threaten, reluctant or even unconscious states. In this paper, we propose a new concept, consent biometrics, which incorporates a consent signature based on human behavioral information. The proposed consent biometrics allows users to register their consent signature. When performing identification, the consent biometrics system will measure whether users consent to access the biometric system as well as perform biometrics identity match. In this way, the consent biometrics system can enhance the security, protect users from unauthorized access to the biometric system and discourage criminals to harm/threaten biometric users to some extent. We also propose two schemes that can realize the consent biometrics idea, and present example designs using face recognition and iris recognition as examples of these two schemes, respectively. The experimental results show that consent biometrics cannot only detect the willingness of users, but also achieve high recognition accuracy.

Introduction

Biometrics identifies/verifies a person using his/her physiological or behavioral characteristics [1], [2], [3], [4]. It is becoming an important ally for law enforcement and homeland security. However, there are still some security and privacy concerns [32], [33], [34].

Some researchers proposed liveness test to prevent spoofing attacks using liveless patterns in the past two decades. In the early 1990s, Daugman [5] emphasized the necessity and importance of liveness detection in iris recognition systems. Reddy et al. [6] introduced a fingerprint liveness detection method which computes the percentage of oxygen in the blood, along with the heart pulse rate to determine the liveness of the enrolled biometric. Nikam et al. [7], [8] proposed ridgelet and curvelet transform-based methods to detect a spoof fingerprint attack. Kanematsu et al. [9] developed a liveness detection method based on variations in the brightness of the iris pattern. Wei et al. [10] proposed three ways using different characteristics of iris texture to detect a fake iris.

However, the liveness test can only detect if the biometric data is acquired alive, which is not enough. Users may access the system under threaten, reluctant or even unconscious states, as a result, criminals can intimidate or even harm the user to use his/her biometric patterns to get into a biometric system. For example, Fig. 1 shows a Youtube video describing the above situation: the attacker knocked the user out and got invalid access using biometric trait of the unconscious user. To prevent such event, it is desirable to have a biometric system that can detect if the user willingly presents the biometric for access. Biometric authentication devices which can detect the willingness of people would be of great help in enhancing security and reducing crimes.

Willingness or intention is defined in the theory of reasoned action (TRA) [26] as “the amount of effort one is willing to exert to attain a goal”. The connection between willingness/intention and behavior is still a challenging research topic. Many research results suggest that adding terms regarding the specific way in which the behavior may be performed can increase the likelihood that intentions to engage in behaviors. In another word, if we know more detailed information about a certain behavior (for example, how the behavior is performed? When and where the behavior will take place?), it is possible to predict behavior from a known willingness/intention. However, it is very challenging to detect certain willingness from a set of behaviors reversely due to the following reasons: first, the behaviors from the same intention would be varied from one person to another, due to their different habits, cultures, ages and education backgrounds, etc. Second, even for the same person, behaviors from the same intention may be still varied a lot, which depends on his/her mood, state of health, etc.

In order to minimize the ambiguity of willingness test, in this research, we propose the concept consent biometrics, which allows the user to enroll his/her consent signature in the system. The consent signature is a definite set of behaviors known by both the recognition system and the test subjects. Instead of detecting the possible connection between a complex behavior set and certain willingness, we detect the consent of each subject based on whether the pre-defined behavior set/consent signature is performed.

The rest of this paper is organized as follows. Section 2 proposes the consent biometrics concept. Section 3 gives an example design of consent face recognition and its related experimental results. Section 4 introduces an example design of consent iris recognition and experimental results. Finally, conclusions and discussions are presented in Section 5.

Section snippets

New concept: Consent biometrics

In this paper, we propose to incorporate consent detection into biometrics, which is called “consent biometrics”. The consent signature can be used to confirm the consent of users, which can be extracted from the user’s physiological or behavior features to enhance security and protect the user against unwillingly access. Consent biometric system is designed to distinguish whether the user is truly willing to do the transaction by checking the extracted consent signature. The consent signature

An example design for combinational consent biometrics using face recognition

Face recognition has been a very active topic for biometrics, computer vision, social interaction, etc. Several successful face recognition systems and algorithms have been proposed during the past two decades [27], [28], [29], [30].

The diagram of the proposed design is shown in Fig. 3. The system acquires a video of sequential facial expressions of a user and the segmentation module segments the face area out. The facial expression sequence is extracted as the consent signature and

An example design for incorporating consent biometric using iris recognition

In this section, we present an example design of incorporating consent biometric scheme using iris recognition. Iris recognition is one of the most accurate and reliable methods of positive identification [10]. In recent years, several methods for iris recognition have been developed [8], [9], [10], [11], [12], [13], [14], [35], [38], [39], [40].

Our proposed system combines the video-based consent detection and multimodal score fusion of iris recognition outputs. The diagram of the proposed

Conclusions and discussions

In this paper, we propose the consent biometrics concept, which can enhance the security, protect users from unauthorized access to the biometric system and discourage criminals to harm/threaten biometric users to some extent. In addition, compared to traditional biometric systems, the consent biometric system can provide additional information for recognition, which can improve the recognition accuracy in some scenarios in spite of a little inconvenience for users.

We propose two consent

Acknowledgement

The author would like to thank Matt Blair for his help and all the subjects contributing to our database.

Kai Yang received the B.S. degree in computer science from Beijing University of Posts and Telecommunications, Beijing, China, in 2009, and the M.S. degree in electrical and computer engineering from Indiana University-Purdue University Indianapolis, Indianapolis, in 2011, where he is currently working toward the Ph.D. degree in the Biometrics and Pattern Recognition Laboratory, Electrical and Computer Engineering Department. His research interests include image processing, biometrics, and

Reference (40)

  • K. Bowyer et al.

    Image understanding for iris biometrics: a survey

    Comput. Vision Image Understanding

    (2008)
  • C. Belcher et al.

    SIFT approach to iris recognition

    Opt. Lasers Eng.

    (2009)
  • I. Ajzen

    The theory of planned behavior

    Organ. Behav. Hum. Dec. Processes

    (1991)
  • P.V. Reddy et al.

    Approach for biometric devices

    IEEE Trans. Biomed. Circuits Syst.

    (2008)
  • S.B, Nikam, S. Agarwal, Fingerprint anti-spoofing using ridgelet transform, in: biometrics: theory, applications and...
  • S.B, Nikam, S. Agarwal, Fingerprint liveness detection using curvelet energy and co-occurrence signatures, in: computer...
  • J. Daugman

    Demodulation by complex-valued wavelets for stochastic pattern recognition

    Int. J. Wavelets Multiresolution Inf. Process.

    (2003)
  • M, Kanematsu, H. Takano, K. Nakamura, Highly reliable liveness detection method for iris recognition, in: SICE, Annual...
  • Z. Wei, X. Qiu, Z. Sun, T. Tan, Counterfeit iris detection based on texture analysis, in: pattern recognition, ICPR...
  • J. Daugman

    How iris recognition works

    IEEE Trans. Circuits Syst. Video Technol.

    (2004)
  • J. Daugman

    New methods in iris recognition

    IEEE Trans. Syst. Man Cybern. Part B

    (2007)
  • N. Sudha et al.

    Iris recognition on edge maps

    IET Comput. Vision

    (2009)
  • Z. Zhou et al.

    Transforming traditional iris recognition systems to work in nonideal situations

    IEEE Trans. Ind. Electron.

    (2009)
  • Y. Du et al.

    Scale invariant Gabor Descriptor-based noncooperative iris recognition,

    EURASIP J. Adv. Signal Process.

    (2010)
  • L. Masek et al.

    Matlab Source Code for a Biometric Identification System Based on Iris Patterns. The School of Computer Science and Software Engineering

    (2003)
  • Y. Du et al.

    Video-based noncooperative iris image segmentation

    IEEE Trans. Syst. Man Cybern. Part B Cybern.

    (2011)
  • D. Lowe

    Distinctive image features from scale- invariant keypoints

    Int. J. Comput. Vision

    (2004)
  • C, Harris, M. Stephens, A combined corner and edge detector, in: Proc. Fourth Alvey Vision Conference,...
  • R. Snelick et al.

    Large-scale evaluation of multimodal biometric authentication using state-of-the-art systems

    IEEE Trans. Pattern Anal. Mach. Intell.

    (2005)
  • Cited by (0)

    Kai Yang received the B.S. degree in computer science from Beijing University of Posts and Telecommunications, Beijing, China, in 2009, and the M.S. degree in electrical and computer engineering from Indiana University-Purdue University Indianapolis, Indianapolis, in 2011, where he is currently working toward the Ph.D. degree in the Biometrics and Pattern Recognition Laboratory, Electrical and Computer Engineering Department. His research interests include image processing, biometrics, and pattern recognition. Mr. Yang was the recipient of the IUPUI Fellowship Award in 2009 and 2011.

    Eliza Yingzi Du is an Associate Professor with the Department of Electrical and Computer Engineering at Indiana University Purdue University Indianapolis (IUPUI). She is the Director of the Biometrics and Pattern Recognition Laboratory. From September 2003 to July 2005, she was an Assistant Research Professor with the Electrical Engineering Department at the United States Naval Academy. Her research interests include image processing, pattern recognition, and biometrics. Her research has been funded by the Office of Naval Research, National Institute of Justice, Department of Defense, National Science Foundation, Indiana Department of Transportation, and several IUPUI internal grants. She received the Office of Naval Research (ONR) Young Investigator award in 2007, the Indiana University Trustee Teaching Award in 2009, the Supervisor of the Year Award at IUPUI in 2009, and the Best Paper Award with her students in IEEE Workshop on Computational Intelligence in Biometrics: Theory, Algorithms, and Applications in 2009. She is a member of the honor societies Tau Beta Pi, and Phi Kappa Phi.

    Zhi Zhou received the BS degree in electrical engineering from Beijing University of Technology, China in 2005. He received the MS degree in electrical and computer engineering from Indiana University-Purdue University Indianapolis (IUPUI) in 2008. Currently, he is a PhD student at IUPUI. His research interests include image processing, biometrics and pattern recognition. He was a Starr Fellow in 2007 and received Best Paper Award in IEEE Workshop on Computational Intelligence in Biometrics: Theory, Algorithms, and Applications in 2009.

    View full text