Real-time Detection of Vehicle Drivers’ Distractions using YOLOv11

Distracted driving is one of the main causes of traffic accidents and poses serious threats to safety of drivers, other road users and public property. Existing methods for detecting distracted driving often suffer from low detection accuracy, poor real-time performance, and high computational demand. This paper proposes a distracted driving behavior detection algorithm based on the latest YOLOv11 object detection model. The YOLOv11's is exhibits enhanced feature extraction capabilities and lightweight network design, which could identify distracted driving behaviors in a highly efficient manner. The study focusses on many different scenarios, such as phone usage, drinking and smoking while maintaining detection accuracy, by aligning with realistic driving scenarios, so as to facilitate development of a more suitable for in-vehicle driver monitoring systems. This study is expected to contribute towards enhanced road traffic safety through efficient detection of distracted driving events, while providing a sound theoretical foundation and technical support for a comprehensive driver detection system.