Chinchilla, R., & Harris, H. J. (2011, June), Facial Recognition System Screening Evaluation Methodology for Complexion Biases  Paper presented at 2011 ASEE Annual Conference & Exposition, Vancouver, BC. 10.18260/1-2--17978

\bibitem{asee_peer_17978}
  Chinchilla, R., \& Harris, H. J. (2011, June), \emph{Facial Recognition System Screening Evaluation Methodology for Complexion Biases}
  Paper presented at 2011 ASEE Annual Conference \& Exposition, Vancouver, BC.
  10.18260/1-2--17978

Rigoberto Chinchilla and Harold Jay Harris.  "Facial Recognition System Screening Evaluation Methodology for Complexion Biases".  2011 ASEE Annual Conference & Exposition, Vancouver, BC, 2011, June.  ASEE Conferences, 2011.  https://peer.asee.org/17978 Internet.  30 Apr, 2024

\bibitem{asee_peer_17978}
  Rigoberto Chinchilla and Harold Jay Harris.
  "Facial Recognition System Screening Evaluation Methodology for Complexion Biases".
  \emph{2011 ASEE Annual Conference \& Exposition, Vancouver, BC, 2011, June}.
  ASEE Conferences, 2011.
  https://peer.asee.org/17978 Internet.  30 Apr, 2024

@INPROCEEDINGS{asee_peer_17978
author = "Rigoberto Chinchilla and Harold Jay Harris"
title = "Facial Recognition System Screening Evaluation Methodology for Complexion Biases"
booktitle = "2011 ASEE Annual Conference \& Exposition"
year = "2011"
month = "June"
address = "Vancouver, BC"
publisher = "ASEE Conferences"
note = {https://peer.asee.org/17978}
number = {10.18260/1-2--17978}
}

TY  - CPAPER
AB  -  Facial Recognition System Screening Evaluation Methodology for Complexion Biases      Over the years, facial recognition systems (FRS) have come under scrutiny fromwatchdog groups who voice their complaints concerning the potential existence of FRSbiases towards certain cultures of people when deployed in security screening situations.Psychology has used the term “other-race effect” to describe this occurrence in humans.Researchers of FRS have used the theory behind the other-race effect to describe theoccurrence of bias associated with algorithms tested during the Face Recognition VendorTest of 2006. In this paper, we develop a scientific methodology based on the factors ofillumination, distance, and angle to evaluate whether or not an operating biometricsystem exhibits a significant bias towards a light complexion 3D facial model over a darkcomplexion 3D facial model. The paper also explains how to integrate this methodologywithin an educational curriculum either in DOE (Design of Experiments) or Biometricscourses. Commercial face recognition (SDK) software has been selected to test themethodology. Using classical DOE techniques, ANOVA and multiple regressionanalysis, this research shows step by step how to assess if biases are present in a facerecognition biometric system toward two different skin complexions. The methodologycan be equally used to test face recognition systems biases toward other racecharacteristics different than the color of the skin. This methodology can easily beextended to include more factors related with the biometric software system in theassessment. A detailed laboratory guide for the students has been developed in theprocess.Index Terms – Biometrics, Facial Recognition, Other-race Effect, Ethics, Minorities
AU  - Rigoberto Chinchilla
AU  - Harold Jay Harris
CY  - Vancouver, BC
DA  - 2011/06/26
PB  - ASEE Conferences
TI  - Facial Recognition System Screening Evaluation Methodology for Complexion Biases
UR  - https://peer.asee.org/17978
DO  - 10.18260/1-2--17978
ER  -