| dc.description.abstract | Drunkenness is now often regarded as one of society’s most serious issues. The majority of road accidents
are caused by drunk driving. Although several solutions have been presented, such as using an Alcohol
Breath Tester (ABT) to check or predict sobriety based on functional condition after alcohol consumption, an automated system to detect sobriety candidates using facial image analysis is needed. This thesis
focuses on proposing a methodology based on evaluating a facial thermal infrared image, adding noise
and flters for augmentation, and determining intoxication using various machine learning algorithms.
In drunkenness detection, most research focus on using RGB image of facial expression like eye sate, head
position, using BAC level or functional state indicators. Sometimes it is not trusty when attempting to
predict on certain people who have certain facial feature patterns that the machine learning algorithm
learned to be a factor of drunkenness.
The combination of using the thermal infrared image with some noise and flter then predicting by
optimized Convolutional Neural Network (CNN) model approach 93% on accuracy proves the efciency
as well as the feasibility of the proposed method. However, because it is unclear how precise such a
system may be, this research will concentrate in part on assessing the solution’s possible viability in a
real-world setting | en_US |
