Computer scientists have published a paper detailing how two algorithms could be used in conjunction with thermal imaging to scan for inebriated people in public places.
In the paper, published in the International Journal of Electronic Security and Digital Forensics, Georgia Koukiou and Vassilis Anastassopoulos of the University of Patras in Greece detail two different algorithms they are working on that focus on data gathered from a subjectâ€™s face â€” alcohol causes blood-vessel dilation at the skinâ€™s surface, so by using this principle as a starting point the two began to compare data gathered from thermal-imaging scans.
One algorithm compares a database of these facial scans of drunk and sober individuals against pixel values from different sites on a subjectâ€™s face. A similar method has been used in the past to detect infections, such as SARS, at airports â€” though a study carried out at the time of the 2003 outbreak warned, â€œalthough the use of infrared instruments to measure body surface temperatures has many advantages, there are human, environmental, and equipment variables that can affect the accuracy of collected data.â€
A second algorithm is used to map out the different areas of the face.