Facial Recognition (CYBR 650, Week 8)

Biometric data including facial recognition is being scooped up by companies including Facebook and Google.  Although this technology is not new, since Casino operators have long used software to try to identify card cheats, it is becoming widespread and far more accurate due to advances in software, hardware, and higher quality, less expensive cameras.

There are two predominant approaches to facial recognition.  One is geometric, which is based on the distances and facial features, such as the distance between eyes.  This is known as “feature based” facial recognition.  Then there is “photometric” which is a “view based” facial recognition system.  Photometric facial recognition relies on reflected light from different facial features as captured by camera and cataloged by computer systems.

Disadvantages of facial recognition systems are that they require high quality cameras and good lighting for the most part.  They can be easily avoided by disguises and cosmetic makeup.  FedTech recognizes four limitations of facial recognition, being:

  1.   Image quality
    2. Image size
    3. Face angle
    4. Processing and storage

Facial recognition is being used by law enforcement to track the movements of suspects on watchlists in public places including airports and has been used to years by the gaming industry to identify card counters and other undesirable customers in casinos.  Facebook is using facial recognition in photographs to identify Facebook members to make it easier to “tag” friends in posted photos.

Facebook explains that they “currently use facial recognition software that uses an algorithm to calculate a unique number (“template”) based on someone’s facial features, like the distance between the eyes, nose and ears. This template is based on your profile pictures and photos you’ve been tagged in on Facebook. We use these templates to help you tag photos by suggesting tags of your friends.”  The Washington Post reported that the Facebook facial recognition software is 97.25% accurate.


Limitations of Facial Recognition Technology. (2013, Nov. 22). Retrieved July 30, 2017, from http://www.fedtechmagazine.com/article/2013/11/4-limitations-facial-recognition-technology

Facebook. (2017). How does Facebook suggest tags? Retrieved July 30th, 2017 from https://www.facebook.com/help/122175507864081

What happens when facial recognition tools are available to everyone. (n.d.). Retrieved May 20, 2016, from https://www.washingtonpost.com/news/innovations/wp/2015/12/23/what-happens-when-facial-recognition-tools-are-available-to-everyone/