Breakthrough Algorithm Detects Driver Drowsiness

A new study reveals that machine learning and deep learning algorithms can effectively detect drowsy drivers, potentially reducing road accidents. The breakthrough approach achieved nearly 100% accuracy in identifying drowsiness, paving the way for safer driving.

Researchers from universities in the United Arab Emirates and Algeria have made significant advancements in driver safety by developing an algorithm that detects driver drowsiness with unprecedented accuracy. The innovative approach leverages machine learning (ML) and deep learning (DL) algorithms to prevent accidents caused by sleepy driving.

According to the National Highway Traffic Safety Administration, drowsiness contributes to about 100,000 road accidents annually in the United States, resulting in 1,500 deaths and 70,000 injuries. This new research aims to curtail these alarming statistics by using technology to warn drivers of their drowsiness, ensuring timely interventions.

“Detecting driver drowsiness [has] become an important task that necessitates an automated system to detect and prevent these adverse outcomes early on,” Saad Harous, a professor of computer science at the University of Sharjah, said in a news release.

The study’s significant innovation lies in its use of electroencephalography (EEG), widely regarded as the “gold standard” for drowsiness detection due to its efficiency and reliability. Traditional EEG drowsiness detection systems, however, have faced challenges in determining optimal preprocessing parameters.

This study is pioneering in its application of an optimization algorithm to find these parameters, resulting in high accuracy and reduced training time.

The research, published in the journal Biomedical Signal Processing and Control, demonstrates the successful integration of convolutional neural networks (CNN) with ML classifiers, creating a deep hybrid learning model.

The algorithm’s mean accuracy score saw an increase from 91% to 95% after optimization, and further improvements raised it to 97%. Most impressively, the hybrid model achieved a near-perfect accuracy rate of 99.9% with a substantial reduction in training time.

“Finally, and most importantly, the use of CNN-SVM classifier achieved the highest average accuracy of 99.9%, and the training time has been reduced to a shallow value,” the authors noted in the study.

The practical implications of this research are vast. Integrating such a system into vehicles could drastically reduce the number of accidents caused by drowsy driving.

Harous emphasized the societal impact, adding, “It can have a big impact on society if the system is adopted by the transportation authority.”

Looking ahead, the researchers are considering practical applications of their algorithm, such as embedding it in a camera or mobile device mounted on the car dashboard. Despite the compelling results, they have yet to receive investment interest from the industry to bring the project to market.

“[W]e have not yet received any emails or invitations from industries willing to invest in the project or requesting us to present our work, though we have achieved the highest accuracy in less time compared to other works,” Harous added.

The research represents a promising step toward safer roads and highlights the potential of artificial intelligence in solving real-world problems. As the team continues to refine their work and seek partnerships, the dream of nearly eliminating accidents caused by drowsy driving edges closer to reality.