Driver Distraction Detection

  • Unique Paper ID: 177870
  • Volume: 11
  • Issue: 12
  • PageNo: 1603-1614
  • Abstract:
  • Injury and fatality figures from road accidents across the globe highlight driver distraction as one of the most dangerous factors. Increased mobile phone usage and vehicle infotainment systems distract drivers, making it crucial to guard and retain their attention. The objective of this research is to efficiently manage the ever-increasing burden of driver distraction in real-time systems by using computer vision and machine learning. This work aims to build a distraction detection model for a driver with a dashboard camera that monitors various driving activities and classifies them into different categories of distraction such as phoning, eating, texting. The model uses convolutional neural network (CNN) architecture to identify patterns within a labeled input dataset to classify ‘distracted’ and ‘attentive’ states. Special attention is given to accuracy while also reducing system lag for real-time implementations. With some fine-tuning, the developed strategy could be seamlessly integrated with the modern ADAS. Competitive accuracy results related to the variety of illumination and viewpoint angles during observation highlight the advanced performance of the provided system in dealing with type of distraction classification tasks. This research paves the way towards more secure modes of transportation through providing an effective way to manage driver behavior to avert distraction-related accidents.

Cite This Article

  • ISSN: 2349-6002
  • Volume: 11
  • Issue: 12
  • PageNo: 1603-1614

Driver Distraction Detection

Related Articles