Dashboard-camera-based Safe Driving Support System

Distracted driving poses a significant and widespread threat to road safety, resulting in severe accidents and loss of lives. Detecting driver distraction promptly is imperative for enhancing road safety. While various methodologies and technologies have been explored to address this issue, we present an innovative, cost-efficient, non-intrusive, and lightweight Safe Driving Support System (SDSS) that utilizes dual dashboard cameras. In addition to conventional driver gaze tracking, our system considers broader aspects, including monitoring the road environment and pedestrian safety. Our study comprises two primary modules: driver distraction detection and pedestrian safety. The driver distraction detection module evaluates distraction by analyzing the driver's gaze direction and the position of pedestrians on the road. Through extensive experimentation, we determined that one particular strategy, strategy B, outperformed others, achieving an 88.76% accuracy rate for Strictly Correct Estimation Rate (SCER) and a 96.07% accuracy rate for Loosely Correct Estimation Rate (LCER). The pedestrian safety module assesses the risk level of pedestrians based on lane lines and their relative positions. Together, these modules offer an early detection and mitigation solution for the leading causes of accidents: driver distraction and pedestrian risk. In summary, our SDSS provides a comprehensive and cost-effective approach to enhancing road safety by addressing driver distraction and pedestrian safety. The results from our experiments demonstrate the system's effectiveness in detecting and mitigating potential hazards on the road, contributing to a safer transportation environment and the prevention of accidents.