MDLNN Approach for Alcohol Detection using IRIS

In this study, a novel approach to Analyzing Near-Infrared (NIR) iris video frames to estimate behavioral curves is discussed. A Fitness for Duty system can employ this technique (FFD). The study aims to ascertain how the Central Nervous System (CNS) is affected by outside elements including alcohol, drugs, and tiredness. The purpose is to examine the representation of this in terms of iris and pupil movements, and behavior that can be observed and explores the possibility of recording these changes through the use of NIR cameras. The behavioral analysis revealed significant differences in pupil and iris behavior, which can be used to determine if an employee is "Fit" or "Unfit". The best results clearly distinguished between participants who were drunk, high, or sleeping, with an overall accuracy of 74.0% for the "Fit" class and 75.5% for the "Unfit" class. The Gradient Boosted Machine and Multi-Layer Perceptron produced the most favorable outcomes. These results present iris-capturing devices as novel applications.