Instances of wrong-way driving pose a substantial risk to road safety. The YOLOv5 (You Only Look Once version 5) deep learning framework is used in our study to provide a new solution for wrong-way vehicle identification. Traditional detection methods frequently fall short with respect to precision and instantaneous execution. Our solution makes use of YOLOv5's capabilities to detect and locate automobiles travelling in the incorrect direction on public roads. For model training and fine-tuning, a broad dataset of annotated traffic footage is assembled. Extensive tests show that our technology detects wrong-way cars with high accuracy and few false positives. Furthermore, the model is capable of real-time processing, making it suited for use in traffic safety applications. This concept offers a significant improvement in road safety technology, with the potential to improve traffic management and minimise wrong-way driving incidents. Deep learning and real-time video analysis together provide a viable option for enhancing overall road safety.
CentraSense: Proactive Vehicle Direction Monitoring
15.03.2024
3516642 byte
Aufsatz (Konferenz)
Elektronische Ressource
Englisch
British Library Conference Proceedings | 1996
|