Related Experiment Video
Updated: May 25, 2025
07:15
Tactile Vibrating Toolkit and Driving Simulation Platform for Driving-Related Research
Published on: December 18, 2020
4.4K
An intelligent network framework for driver distraction monitoring based on RES-SE-CNN
Jichong Lei1, Zining Ni2, Zhiqiang Peng2
1School of Safety and Management Engineering, Hunan Institute of Technology, Hengyang, Hunan, China. leijichong@gmail.com.
Scientific Reports
|February 26, 2025
Summary
This study introduces an intelligent driver monitoring system using the RES-SE-CNN model to detect distracted driving. The model achieves high accuracy and is suitable for real-time application in vehicles.
Area of Science:
- Artificial Intelligence
- Computer Vision
- Road Safety
Background:
- Increasing vehicle numbers heighten road complexity and accident risk.
- Distracted driving is a major human factor contributing to traffic accidents.
- Real-time monitoring and warnings for driver distraction are crucial.
Purpose of the Study:
- To propose an intelligent driver state monitoring methodology.
- To develop a system for real-time detection and warning of distracted driving behaviors.
- To evaluate the proposed model against established deep learning architectures.
Main Methods:
- Development of a driver state monitoring methodology using the RES-SE-CNN model.
- Comparative analysis against VGG19, DenseNet121, and ResNet50 models.
- Utilizing transfer learning for application in real-world driving environments.
Main Results:
- The RES-SE-CNN model demonstrated superior performance in detecting driver distraction.
- Achieved a correct recognition rate of 97.28%.
- The model exhibits lower memory occupancy, facilitating deployment on in-vehicle systems.
Conclusions:
- The RES-SE-CNN model is highly effective for intelligent driver distraction monitoring.
- The proposed model shows significant potential for real-world application in enhancing road safety.
- Transfer learning enables efficient deployment of driver monitoring systems.
