Sign in Subscribe
Traffic accidents

Enhanced Driver Behavior Detection in IoMT Environments using Hybrid LSTM-GRU Model

Strategic Insights

— 1 min read

Researchers from King Saud University in Riyadh, Saudi Arabia, have developed an advanced framework for real-time driver behavior detection using wearable sensors and deep learning models. The study, published in IEEE Access, highlights the alarming rate of traffic accidents attributed to driver distractions, emphasizing the need for effective safety measures. The proposed system integrates Shimmer3 EMG sensors and a Raspberry Pi device to collect and process bio-signal data, employing a hybrid Long Short-Term Memory (LSTM) and Gated Recurrent Unit (GRU) model for classification.

Key Takeaways:

  • The global rise in traffic accidents is significantly attributed to driver distractions, including cell phone use, eating, drinking, and interacting with in-vehicle systems.
  • The proposed system utilizes wearable sensors and deep learning models to detect and analyze driver behaviors in real-time, improving road safety through enhanced driver behavior detection.
  • The research introduced a hybrid LSTM-GRU model for classification, achieving a classification accuracy of 83.40% and 77.90% for two real-life public datasets.
  • The study underscores the effectiveness of combining LSTM and GRU layers in capturing complex temporal dependencies, offering a scalable and reliable solution for enhancing road safety.
  • The proposed system includes three key modules: data acquisition, hybrid LSTM-GRU training, and real-time analysis and feedback.
  • Experimental evaluations demonstrated the system's ability to detect non-aggressive and aggressive driving behaviors.

Statistics:

  • The classification accuracy of the proposed hybrid model was 83.40% and 77.90% for the first and second datasets, respectively.
  • The system utilizes wearable Shimmer3 EMG sensors and a Raspberry Pi device to collect and process bio-signal data.
  • The proposed framework includes three key modules: data acquisition, hybrid LSTM-GRU training, and real-time analysis and feedback.

Sources:

  • Enhanced Multi-Class Driver Behavior Detection in IoMT Environments Using Hybrid LSTM-GRU Model. IEEE Access, 2025,13():142379-142396. (IEEE Access - http://ieeexplore.ieee.org/servlet/opac?punumber=6287639)
  • Our news editors report that additional information may be obtained by contacting Hussain AlSalman, Department of Computer Science, College of Computer and Information Sciences, King Saud University, Riyadh, Saudi Arabia.

Read more