AI Drive Assist: Enhancing Road Safety through Advanced Road Sign Detection and Driver Alerts

Abstract

Road accidents claim approximately 1.35 million lives annually, as highlighted by the World Health Organization, making them a leading cause of death worldwide. With this sobering statistic in mind, addressing road safety concerns through advanced AI technologies becomes increasingly imperative. This research paper highlights practical implications for both road safety enhancement and the advancement of artificial intelligence technology within the automotive industry. The proposed model, "AI-drive assist," investigates the effectiveness of an AI-driven driving assistant system in improving road safety. By providing real-time auditory alerts to drivers, the driving assistant system facilitates a better understanding of road conditions and encourages the development of safer driving habits. The methodology entails the utilization of the You Only Look Once (YOLO)v8 model within a ResNet-50 CNN framework, allowing for the efficient extraction of relevant information from input photographs. Through rigorous evaluation, the system achieves an impressive precision-recall rate of 94% in identifying various road signs, indicating its potential to enhance driver awareness and promote compliance with traffic regulations. Additionally, data augmentation techniques are employed to diversify the training dataset, further improving accuracy and robustness. The findings of this study underscore the significant impact of AI technologies on promoting safer driving practices. Overall, this study contributes to the ongoing discourse on road safety improvement and demonstrates the tangible benefits of integrating AI technologies into driving assistance systems.