Presented at The 8th International Conference on Information Technology and Applications (ICITA 2013), Sydney Australia, July 1 - July 4 2013.
The purpose of this paper is to illustrate the automatic detection of biometric transaction times using hand geometry as the modality of interest. Video recordings were segmented into individual frames and processed through a program to automatically detect interactions between the user and the system. Results include a mean enrollment time of 15.860 seconds and a mean verification time of 2.915 seconds.
Axa Assurance Maroc - Insurer Innovation Award 2024
(2013) Automatic Detection of Biometrics Transaction Times
1. BIOMETRICS LAB
Biometric Standards, Performance and Assurance Laboratory
Department of Technology, Leadership and Innovation
AUTOMATIC DETECTION OF
BIOMETRIC TRANSACTION
TIMES
MICHAEL BROCKLY
STEPHEN ELLIOTT PH.D.
2. HAND GEOMETRY
• Measures length,
width, and thickness
of hand [1]
• Engages 1:1 matching by entering a
Personal Identification Number (PIN)
[1]
3. USES
• Joins a PIN number with the security of
biometric verification
• Commonly used in time and attendance
and access control
• Hand geometry has proven to be very
popular in time and attendance recording
[2]
4. BENEFITS
• Hand geometry functions as a medium
cost system with fast computational
speeds, low template size, and good
ease of use [3]
• The convenience of hand geometry
stems from the fact that users cannot
lose or forget their biometric credential
[4]
5. TIME ON TASK
• Computational speed is always a
primary concern
• Slow throughput times may eliminate the
cost savings proposed by device
installation
• Higher costs are associated with a
higher time to acquire or process a
biometric sample [5]
6. VIDEO CODING
• Previous studies suggest video
recording in order to capture subject time
on task [6]
• Time consuming process to manually
record timing data
• Potential for errors and inconsistencies
7. INTERRATER RELIABILITY
• Represents the degree to which the
ratings of different judges are
proportional when expressed as
deviations from their means [7]
• Not all video coders will report the same
result
8. OPERATIONAL TIMES
• Previous research has suggested
models for biometric transaction times
• Biometric transaction time includes:
– Subject interaction time
– Biometric subsystem processing time
– Biometric subsystem decision time
– External control access time
12. CAMERA
• Logitech HD Pro C910
Webcam
– 1080p recording
• Used to video record
interaction changes on
hand geometry device
13. SETUP
• Camera placed 24 cm above
hand geometry machine
• Device placed 90 cm above
ground level
14. EXPERIMENT
• Hand geometry data was collected as
part of a larger multi-modal study
• This data collection included 35 subjects
• Other modalities collected include
fingerprint, iris, face, signature, and palm
vein
16. USES
• An automated tool was created to
analyze the videos
• Analyzes videos to 15 frames per
second
• Detects light changes on device as pixel
color thresholds are crossed
• Writes results without human coder
20. BIOMETRICS LAB
Biometric Standards, Performance and Assurance Laboratory
Department of Technology, Leadership and Innovation
TRANSACTION TIME
USE CASE – HAND
GEOMETRY
32. CONFLICTING TERMINOLOGY
• Along with the model, we include specific
terminology and emphasize the linkages
between the two versions
33. TRANSACTION
• The sequence of attempts to the system
on the part of the user for the purpose of
enrollment, verification or identification
• This definition follows ISO/IEC FCD
19795-1’s definition of a transaction
34. ATTEMPT
• The submission of one (or a sequence
of) biometric samples to the system on
the part of the user
– One or more attempts as allowed by the
biometric system will create one transaction
• This definition follows ISO/IEC FCD
19795-1’s definition of an attempt
35. PRESENTATION
• The submission of a single biometric
sample to the system on the part of the
user
– One or more presentations as allowed by the
biometric system will create one attempt
• This definition follows ISO/IEC FCD
19795-1’s definition of a presentation
36. INTERACTION
• The action(s) that take place within a
presentation
– One or more interactions will create one
presentation
• This definition conflicts with ISO/IEC
FCD 19795-1’s definition as “a sequence
of transactions”
43. BENEFITS OF AUTOMATIC
CODING
• Eliminates need for manual video coding
• Video coding is a time consuming task
and has potential for errors
• Goal is to create a consistent measure of
biometric transactions
44. LESSONS LEARNED
• Experimental test conditions are not
always stable
– Due to cameras being moved/bumped, they
will not always be in the same location
• Original version of software did not take
this into account
• Second version allowed the area of
interest to be selected based on a frame
of the video
45. RELATION TO HBSI
• This experiment addresses the need to
automate the error detection in the
Human Biometric Sensor Interaction
(HBSI) model
• HBSI is concerned with classifying
correct and incorrect presentations into
quantifiable metrics
47. HBSI
• This philosophy can be duplicated to
record these error metrics
• Ex. 1 If all lights are extinguished and
green light is shown, SPS
• Ex 2. If all lights remain on until system
time out and red light is shown, FTD
48. NEXT STEPS
• Methodology can be replicated for other
modalities as well
• Any system that provides feedback can
be video recorded and analyzed
• Automatically code HBSI error metrics
51. REFERENCES
[1] Sidlauskas, D., Tamer, S., (2007). Hand Geometry Recognition.
Handbook of Biometrics. Springer US. doi: 10.1007/978-0-387-
71041-9_5
[2] Liu, S., & Silverman, M. (2001). A practical guide to biometric security
technology. IT Professional, 3(1), 27–32. Retrieved from
http://ieeexplore.ieee.org/xpls/abs_all.jsp?arnumber=899930
[3] Sanchez-Reillo, R., & Gonzalez-Marcas, A. (2000). Access control
system with hand geometry verification and smart cards. Aerospace
and Electronic Systems Magazine, IEEE, 15(45), 45–48. Retrieved
from http://ieeexplore.ieee.org/xpls/abs_all.jsp?arnumber=82 5671
[4] Tamer, S., Elliott, S., (2009, July) Time and Attendance.
Encyclopedia of Biometrics. Springer US. doi:10.1007/978-0-387-
73003-5_114
52. REFERENCES
[5] Poh, N., Bourlai, T., & Kittler, J. (2010). A multimodal biometric test bed
for quality-dependent, cost-sensitive and client-specific score-level
fusion algorithms. Pattern Recognition, 43(3), 1094–1105.
doi:10.1016/j.patcog.2009.09.011
[6] Bailey, B. P., Konstan, J. a., & Carlis, J. V. (2000). Measuring the
effects of interruptions on task performance in the user interface. 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.00CH37166), 2, 757–762. doi:10.1109/ICSMC.2000.885940
[7] Reliability and Agreement of Subjective Judgments. Journal of
Counseling Psychology, 22(4), 358–376.
[8] Lazarick, R. T., Kukula, E. P., & Elliott, S. J. (2009, July).
Operational Times. Encyclopedia of Biometrics. Springer US.
doi:10.1007/978-0-387-73003-5_114
Hinweis der Redaktion
1. Sidlauskas, D., Tamer, S., (2007). Hand Geometry Recognition. Handbook of Biometrics. Springer US. doi: 10.1007/978-0-387-71041-9_5
2. Liu, S., & Silverman, M. (2001). A practical guide to biometric security technology. IT Professional, 3(1), 27–32. Retrieved from http://ieeexplore.ieee.org/xpls/abs_all.jsp?arnumber=899930
3. Sanchez-Reillo, R., & Gonzalez-Marcas, A. (2000). Access control system with hand geometry verification and smart cards. Aerospace and Electronic Systems Magazine, IEEE, 15(45), 45–48. Retrieved from http://ieeexplore.ieee.org/xpls/abs_all.jsp?arnumber=8256714. Tamer, S., Elliott, S., (2009, July) Time and Attendance. Encyclopedia of Biometrics. Springer US. doi:10.1007/978-0-387-73003-5_114
5. Poh, N., Bourlai, T., & Kittler, J. (2010). A multimodal biometric test bed for quality-dependent, cost-sensitive and client-specific score-level fusion algorithms. Pattern Recognition, 43(3), 1094–1105. doi:10.1016/j.patcog.2009.09.011
6. Bailey, B. P., Konstan, J. a., & Carlis, J. V. (2000). Measuring the effects of interruptions on task performance in the user interface. 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.00CH37166), 2, 757–762. doi:10.1109/ICSMC.2000.885940
7. Tinsley, H. E. A., & Weiss, D. J. (1975). Interrater Reliability and Agreement of Subjective Judgments. Journal of Counseling Psychology, 22(4), 358–376.
8. Lazarick, R. T., Kukula, E. P., & Elliott, S. J. (2009, July). Operational Times. Encyclopedia of Biometrics. Springer US. doi:10.1007/978-0-387-73003-5_114
This process begins when the PIN is enteredFor hand geometry verification only one attempt is given in the transaction.
In verification, one attempt contains one presentation.
Hand geometry enrollment is made up of 3 presentations of sufficient quality.
Signified by the lights on the hand geometry machine changing color. This may happen many times within a presentation to the systemInteraction occurs between the subject and the system.Instructions should be provided to the subject before the first interaction begins.
Video coding provides consistency.Bullet 2 is a rehash from the introduction