acm-header
Sign In

Communications of the ACM

ACM TechNews

Aging Faces Could Increase Security Risks


View as: Print Mobile App Share:
A repeat offender at age 30 and 39; the computer was unable to match these as the same person.

The accuracy of automatic facial-recognition systems begins to drop if the images of a person were taken more than six years apart, due to natural changes that occur to a face as a person ages.

Credit: Anil Jain

Michigan State University researchers led by Anil Jain are studying how facial aging affects the performance of automatic facial-recognition systems and what implications this could have on successfully identifying criminals or determining when identification documents need to be renewed.

The researchers found 99% of facial images can be recognized as many as six years later. However, the results also showed after six years, recognition accuracy begins to decline due to natural changes that occur to the face. The researchers say this decrease in facial-recognition accuracy is person-dependent, as some people age faster than others due to lifestyle, health conditions, environment, or genetics.

The team based its research on two police mugshot databases of repeat criminal offenders with each offender having a minimum of four images acquired over at least a five-year period, for a total of 23,600 images.

From MSUToday
View Full Article

 

Abstracts Copyright © 2017 Information Inc., Bethesda, Maryland, USA


 

No entries found

Sign In for Full Access
» Forgot Password? » Create an ACM Web Account