A facial recognition system is generally thought of as a computer application to identify a person using facial features from an existing image. Now, a new facial recognition system is trying to turn that concept completely around. According to a recent article in SmartPlanet:
Mark Shriver of Pennsylvania State University and his team have been gathering 3D images of hundreds of volunteers and their saliva over the past several months. These photographs have then been used to find over 7,000 points of reference from facial features, then fed in to software which finds similarities between DNA, sex, race and facial features.
Shriver found that 20 genes with only 24 variants were “reliable indicators” of a person’s facial shape — including eye shape, nose, forehead, cheeks and chin. Once plotted, these features could then be constructed through the program and 3D-printed to produce a copy of the person’s face.
The researchers said:
“Results on a set of 20 genes showing significant effects on facial features provide support for this approach as a novel means to identify genes affecting normal-range facial features and for approximating the appearance of a face from genetic markers.”
Based on this discovery, Shriver and his team want to make the system even more reliable, and now aim to plot 30,000 different points on each face.
Thousands of crimes go unsolved every year, and facial recall by witnesses is far from an exact science. However, if DNA could be gathered at a crime scene — whether hair, saliva or fluid — and then a reliable model of the suspect’s face could be produced, more criminals could be caught, and there would be fewer wrongly convicted people due to faulty recall of someone’s face