Internet of things based smart framework for the safe driving experience of two wheelers
Gunjan Chhabra,Keshav Kaushik,Pardeep Singh,Gourav Bathla,Ahmad Almogren,Salil Bharany,Ayman Altameem,Ateeq Ur Rehman
DOI: https://doi.org/10.1038/s41598-024-72357-4
2024-09-18
Abstract:Several parameters affect our brain's neuronal system and can be identified by analyzing electroencephalogram (EEG) signals. One of the parameters is alcoholism, which affects the pattern of our EEG signals. By analyzing these EEG signals, one can derive information regarding the alcoholic or normal stage of an individual. Many road accident cases around the world, including drinking and driving scenarios, which result in loss of life, have been reported. Another reason for such incidents is that riders avoid wearing helmets while driving two-wheelers. Many road accident cases involving two-wheelers, including drinking, driving, overspeeding, and nonwearing helmets, have been reported. Therefore, to solve such issues, the present work highlights the features of an intelligent model that can predict the alcoholism level of the subject, wearing of a helmet, vehicle speed, location, etc. The system is designed with the latest technologies and is smart enough to make decisions. The system is based on multilayer perceptron, histogram of oriented gradients (HoG) feature extraction, and random forest to make decisions in real time. The accuracy of the proposed method is approximately 95%, which will reduce the fatality rate due to road accidents. The system is tested under different working environments, i.e., indoor and outdoor, and satisfactory outcomes are observed.