Real-Time Distracted Driver Detection and Alert System for Enhanced Road Safety

Abstract

A method for detecting distracted drivers that uses Convolutional Neural Networks (CNNs) to reliably identify distractions in real-time. There are a lot of accidents and fatalities every year caused by people who aren't paying attention to the road. Conventional detection methods frequently depend on human observation or basic sensor-based techniques, neither of which are always reliable or applicable in actual driving situations. On the other hand, our suggested approach automatically identifies obstacles in in-car camera footage by employing convolutional neural networks (CNNs), a deep learning architecture that is well-suited to image identification applications. The CNN learns to identify and categorize passenger activities with great accuracy by training it on numerous datasets that comprise different types of distractions, such as using a smartphone, eating, grooming, and socializing with other passengers. In addition, we obtain real-time speed by optimizing the CNN architecture and deploying it on embedded devices. Findings from the experiments show that the suggested system can reliably detect instances of distracted driving, which improves road safety and aids in the fight against accidents. A safer and more efficient transportation future is promised by the system's capacity to detect distractions in real-time, which opens possibilities for integration into driverless vehicles and advanced driver assistance systems (ADAS).