The concept of thermal imaging may assist in identifying drunks in airports, bars and other public areas. Vassilis Anastassopoulos and Georgia Koukiou from the Electronics Laboratory of the University of Patros in Greece are generating a software that can impartially determine if an individual has ingested too much alcohol based on the temperature on different portions of the individual’s face.
In their write-up in the International Journal of Electronic Security and Forensics, the researchers explained that this method will avoid the issue of biased views and will enable the law regulation and affiliated authorities to obtain solid proof of inebriety. They elucidate how the two algorithms that they have come up with can identify if an individual has been consuming excess of alcohol through infrared thermal imaging of the individual’s face.
The first methodology comprises calculating pixel values of certain portions of an individual’s face and comparing them to values in a database of images of drunk and sober persons. Knowing that alcohol results in dilation of blood vessels on the skin surface, hot spots in the face can be viewed in the thermal images, thus helping in identifying sober and drunk portions of the face. The same kind of methodology was used at national borders and other places to identify people who are afflicted by a virus like SARS or the flu.
As per the second methodology, experts can note the thermal differences in various facial regions and analyze the values. Enhanced thermal illumination is usually seen on a drunk person’s nose while the forehead remains cooler. This method depends on the algorithm of “understanding” the portions of the face viewed in the image. A sober image of the person will not be needed to recognize if a person has been drinking.
The two methods can be used together to scan persons entering public areas.
Disclaimer: The views expressed here are those of the author expressed in their private capacity and do not necessarily represent the views of AZoM.com Limited T/A AZoNetwork the owner and operator of this website. This disclaimer forms part of the Terms and conditions of use of this website.