1 / 10

The User and Biometric System Uncertainty

Mary Theofanos. The User and Biometric System Uncertainty. International Workshop on Usability and Biometrics June 23- 24, 2008. The field of biometrics has tested, measured and reported performance statistics independent of the user Error Bounds are defined as combination of:

masao
Download Presentation

The User and Biometric System Uncertainty

An Image/Link below is provided (as is) to download presentation Download Policy: Content on the Website is provided to you AS IS for your information and personal use and may not be sold / licensed / shared on other websites without getting consent from its author. Content is provided to you AS IS for your information and personal use only. Download presentation by click this link. While downloading, if for some reason you are not able to download a presentation, the publisher may have deleted the file from their server. During download, if you can't get a presentation, the file might be deleted by the publisher.

E N D

Presentation Transcript


  1. Mary Theofanos The User and Biometric System Uncertainty International Workshop on Usability and Biometrics June 23- 24, 2008

  2. The field of biometrics has tested, measured and reported performance statistics independent of the user • Error Bounds are defined as combination of: • uncertainty resulting from random effect • uncertainty resulting from systematic effect 2

  3. Can we link the systematic uncertainty of fingerprint quality and performance metrics to human factors? Biometric System Factors • Anthropometrics • Affordance • Instructions • Accessibility User Characteristics • Age • Gender • Height • Experience • Ability • Perception 3

  4. Habituation Study found a number of human factors that influence quality Age: Younger participants submit higher quality images than older participants Gender: Women’s images, on average, are of poorer quality than men’s Feedback Habituation has no influence without feedback Older participants left higher quality images over time with feedback

  5. Anthropometrics: Effect of Sensor Height on quality Image quality: All fingers are sensitive to height except Right index finger Thumbs are more sensitive to height than slaps Left slap is more sensitive than right slap Drop in quality from individual thumbprints to simultaneously captured thumbprints Consistently able to provide higher quality prints for work surfaces lower than 42 inches (107 cm)

  6. Anthropometrics: Effect of Sensor Height on Efficiency Time (Efficiency) Counter height of 36 inches (91cm) yields fastest performance Most efficient capture sequence starts with right hand

  7. Instructions are significant factors for performance and quality • Poster Participants: • took significantly longer to complete the 10-print collection process • made significantly more errors • only 56% were able to successfully complete the fingerprinting process • left the poorest quality images using NFIQ. Verbal and video instructions performed equally well

  8. Operators are critical to the acquisition process • Operators are able to assist individuals to overcome the deficits of the instructional materials • With operator assistance 98% of participants were able to successfully complete the fingerprint process Verbal instructions were most preferred by participants

  9. What about counter height and angle of the fingerprint scanner ?

  10. We have demonstrated a link between human factors quality and performance metrics At BCC last fall Jim Wayman stated that: Laboratory results are not a good predictor of “real-world” performance Systematic uncertainty attributable to uncontrolled variability in human factors Improving biometrics will require emphasis on human factors not the purely “technical” aspects

More Related