An intrinsic assessment and comparison of biometric systems through wavelet analysis

Most, if not all, assessments or comparisons of biometric systems focus on extrinsic characteristics; our focus is intrinsic, as we are concerned with the biometric signals and images that underpin each system, and not with how these signals are obtained, matched or interpreted. Through the intrinsic assessment of a biometric system, we expect to answer several policy-related questions, including: How robust is the system? How can the system be refined? To what extent is the system intrinsically prone to error? Likewise, through the intrinsic comparison of two or more biometric systems, we expect to answer questions concerning: To what extent is the system complementary? Can a more robust, hybrid system be identified? As with any assessment or comparative analysis of a number of systems, it is necessary to carry out the analysis within a consistent framework or model. After considering a number of approaches, we employ the wavelet transform technique because of its robustness in modeling most, if not all, of the biometric signals and images and its ability to shed light on the above policy-related questions.

[1]  John Daugman.  High confidence recognition of persons by iris patterns , 2001, Proceedings IEEE 35th Annual 2001 International Carnahan Conference on Security Technology (Cat. No.01CH37186).

[2]  Anil K. Jain,et al.  Integrating Faces and Fingerprints for Personal Identification , 1998, IEEE Trans. Pattern Anal. Mach. Intell..

[3]  John Daugman,et al.  Demodulation by Complex-Valued Wavelets for Stochastic Pattern Recognition , 2003, Int. J. Wavelets Multiresolution Inf. Process..

[4]  Gregory D. Abowd,et al.  The smart floor: a mechanism for natural user identification and tracking , 2000, CHI Extended Abstracts.

[5]  Sudeep Sarkar,et al.  An evaluation of face and ear biometrics , 2002, Object recognition supported by user interaction for service robots.

[6]  Sharath Pankanti,et al.  Fingerprint Minutiae: A Constructive Definition , 2002, Biometric Authentication.

[7]  Nalini K. Ratha,et al.  Automated Biometrics , 2001, ICAPR.

[8]  Brani Vidakovic,et al.  Wavelet Estimation of a Baseline Signal from Repeated Noisy Measurements by Vertical Block Shrinkage , 2002 .

[9]  Sharath Pankanti,et al.  A Prototype Hand Geometry-based Verication System , 1999 .

[10]  John D. Woodward,et al.  Army Biometric Applications: Identifying and Addressing Sociocultural Concerns , 2001 .

[11]  Sharath Pankanti,et al.  On the similarity of identical twin fingerprints , 2002, Pattern Recognit..

[12]  Sharath Pankanti,et al.  On the Individuality of Fingerprints , 2002, IEEE Trans. Pattern Anal. Mach. Intell..

[13]  Sajjad Haider,et al.  A multi-technique approach for user identification through keystroke dynamics , 2000, Smc 2000 conference proceedings. 2000 ieee international conference on systems, man and cybernetics. 'cybernetics evolving to systems, humans, organizations, and their complex interactions' (cat. no.0.