Stephen J. Elliott

Impact of Age Groups on Fingerprint Recognition Performance

Ever since introduction of automated fingerprint recognition in law enforcement in the 1970s it has been utilized in applications ranging from personal authentication to civilian border control. The increasing use of automated fingerprint recognition puts on it a challenge of processing a diverse range of fingerprints. The quality control module is important to this process because it supports consistent fingerprint detail extraction which helps in identification / verification. Inherent feature issues, such as poor ridge flow, and interaction issues, such as inconsistent finger placement, have an impact on captured fingerprint quality, which eventually affects overall system performance. Aging results in loss of collagen; compared to younger skin, aging skin is loose and dry. Decreased skin firmness directly affects the quality of fingerprints acquired by sensors. Medical conditions such as arthritis may affect the users ability to interact with the sensor, further reducing fingerprint quality. Because quality of fingerprints varies according to the user populations ages and fingerprint quality has an impact on overall system performance, it is important to understand the significance of fingerprint samples from different age groups. This research examines the effects of fingerprints from different age groups on quality levels, minutiae count, and performance of a minutiae-based matcher. The results show a difference in fingerprint image quality across age groups, most pronounced in the 62-and-older age group, confirming the work of [7].

Liveness detection of fingerprint based on band-selective Fourier spectrum

This paper proposes a novel method for fingerprint liveness detection based on band-selective Fourier spectrum. The 2D spectrum of a fingerprint image reflects the distribution and strength in spatial frequencies of ridge lines. The ridge-valley texture of the fingerprint produces a ring pattern around the center in the Fourier spectral image and a harmonic ring pattern in the subsequent ring. Both live and fake fingerprints produce these rings, but with different amplitudes in different spatial frequency bands. Typically, live fingerprints show stronger Fourier spectrum in the ring patterns than the fake. The proposed method classifies the live and the fake fingerprints by analyzing the band-selective Fourier spectral energies in the two ring patterns. The experimental results demonstrate this approach to be a promising technique for making fingerprint recognition systems more robust against fake-finger-based spoofing vulnerabilities.

The Human–Biometric-Sensor Interaction Evaluation Method: Biometric Performance and Usability Measurements

This paper discusses the human-biometric-sensor interaction (HBSI) evaluation method that uses ergonomics, usability, and sample quality criteria as explanatory variables for the overall biometric system performance. The HBSI method was proposed because of questions regarding the thoroughness of traditional system-level performance evaluation metrics such as the failure-to-acquire (FTA) rate, the failure-to-enroll (FTE) rate, the false-accept rate (FAR), and the false-reject rate (FRR). Data were collected from 85 individuals over three visits that accounted for 25 867 user interactions with three swipe-based fingerprint sensors. The results in this paper revealed that traditional biometric evaluations that focus on system-level metrics are not providing sufficient reporting details regarding the user interaction with the devices. In this paper, the systemic FTA rate of 14.38% was shown to be segmented into three metrics: false interaction (FI), failure to detect (FTD), and concealed interaction (CI). The results show that the FI accounted for 69.05% of the systemic FTA presentations, FTD accounted for 30.71%, and CI accounted for 0.24%. Overall, the HBSI evaluation method and framework for biometric interactions provided new metrics that improve the analysis capabilities for biometric performance evaluations as it links system feedback to the human-sensor interaction, enabling researchers, system designers, and implementers to understand if the issues are the result of the system, the user, both the system and the user, or some other extraneous factor.

An evaluation of fingerprint image quality across an elderly population vis-a-vis an 18-25 year old population

This study evaluated fingerprint quality across two populations, elderly and young, in order to assess age and moisture as potential factors affecting utility image quality. Specifically, the examination of these variables was conducted on a population over the age of 62, and a population between the ages of 18 and 25, using two fingerprint recognition devices (capacitance and optical). Collected individual variables included: age, gender, ethnic background, handedness, moisture content of each index finger, occupation(s), subjects use of hand moisturizer, and prior usage of fingerprint devices. Computed performance measures included failure to enroll, and quality scores. The results indicated there was statistically significant evidence that both age and moisture affected effectiveness image quality of each index finger at /spl alpha/=0.01 on the optical device, and there was statistically significant evidence that age affected effectiveness image quality of each index finger on the capacitance device, but moisture was only significant for the right index finger at /spl alpha/=0.01.

Keystroke Dynamics Verification Using a Spontaneously Generated Password

Current keystroke dynamics applications have tackled the problem of traditional knowledge-based static password verification, but the problem of spontaneous password verification persists. The intent of this study was to examine the predictive strength of typing patterns for spontaneous passwords. The typing patterns of an individual typing at a DELLreg keyboard on a DELL OptiPlexreg GX260 machine were recorded. Variables collected included keystroke press time and keystroke latency. Computed performance measures included false match rates (FMR) and false non match rates (FNMR) at various threshold levels

The effects of human interaction on biometric system performance

This paper discusses the impact of human interaction with biometric devices and its relationship to biometric performance. The authors propose a model outlining the Human-Biometric Sensor Interaction and discuss its necessity through case studies in fingerprint recognition, hand geometry, and dynamic signature verification to further understand the human-sensor interaction issues and underlying problems that they present to the biometric system. Human factors, human-computer interaction and digital human modeling are considered in the context of current and future biometric research and development.

Effects of illumination changes on the performance of Geometrix FaceVision/spl reg/ 3D FRS

This evaluation examined the effects of four frontal light intensities on the performance of a 3D face recognition algorithm, specifically testing the significance between an unchanging enrollment illumination condition (220-225 lux) and four different illumination levels for verification. The evaluation also analyzed the significance of external artifacts (i.e. glasses) and personal characteristics (i.e. facial hair) on the performance of the face recognition system (FRS). Collected variables from the volunteer crew included age, gender, ethnicity, facial characteristics, hair covering the forehead, scars on the face, and glasses. The analysis of data revealed that there are no statistically significant differences between environmental lighting and 3D FRS performance when a uniform or constant enrollment illumination level is used.

The impact of fingerprint force on image quality and the detection of minutiae

It is well documented that many factors affect fingerprint image quality such as age, ethnicity, moisture, temperature and force, although force has only been subjectively measured in the literature. Fingerprint image quality is of utmost importance due to its linear relationship with matching performance. Therefore, the purpose of this research is to show how fingerprint force impacts image quality and the number of detected minutiae. Two experiments are presented in this paper that evaluated fingerprint force levels and the impacts on image quality, number of minutiae detected, and user comfort to provide the community with a quantitative measure for force as it relates to image quality. Four force levels (3, 9, 15, and 21 newtons) were evaluated in the first experiment with results indicating that there was no incremental benefit in terms of image quality when using more than 9 N when interacting with an optical fingerprint sensor. The second experiment investigated the 3-9 N interval with results indicating that the optimal image quality is arrived between a force level is 5-7 N.

On improving interoperability of fingerprint recognition using resolution compensation based on sensor evaluation

The purpose of this paper is the development of a compensation algorithm by which the interoperability of fingerprint recognition can be improved among various different fingerprint sensors. In order to compensate for the different characteristics of fingerprint sensors, an initial evaluation of the sensors using both the ink-stamped method and the flat artificial finger pattern method was undertaken. Then the resulted image resolution was incorporated to the compensation algorithms. This paper proposes Common resolution method and Relative resolution method for compensating different resolutions of fingerprint images captured by disparate sensors. Both methods can be applied to image-level and minutia-level. This paper shows the results of the minutiaelevel compensation. The Minutiae format adhered to the standard format established by ISO/IEC JTC1/SC37. In order to compensate the direction of minutiae in minutia-level, Unit vector method is proposed. The fingerprint database used in the performance evaluation is part of KFRIA-DB (Korea Fingerprint Recognition Interoperability Alliance Database) collected by the authors and supported by KFRIA. Before compensation, the average EER was 8.62% and improved to 5.37% by the relative resolution compensation and to 6.37% by the common resolution compensation. This paper will make a significant contribution to interoperability in the system integration using different sensors.

Investigating the relationship between fingerprint image quality and skin characteristics

This paper reports the correlations between skin characteristics, such as moisture, oiliness, elasticity, and temperature of the skin, and fingerprint image quality across three sensing technologies. Fingerprint images from the index finger of the dominant hand of 190 individuals, were collected on nine different fingerprint sensors. The sensors included four capacitance sensors, four optical sensors and one thermal fingerprint sensor. Skin characteristics included temperature, moisture, oiliness and elasticity, were measured prior to the initial interaction with each of the individual sensors. The analysis of the full dataset indicated that the sensing technology and interaction type (swipe or touch) were moderately and weakly correlated respectively with image quality scores. Correlation analysis between image quality scores and the skin characteristics were also made on subsets of data, divided by the sensing technology. The results did not identify any significant correlations. This indicates that further work is necessary to determine the type of relationship between the variables, and how they impact image quality and matching performance.