Dynamic Traffic Signal Optimization Utilizing Real-Time Lane Analysis with YOLO Framework

For applications in autonomous driving and surveillance, efficient vehicle detection is of significant interest. The YOLO framework marked a paradigm shift in this landscape by becoming synonymous with real-time object detection. In this work, we propose the integration of real-time vehicle detection within a stream of optimizations to the YOLO architecture, which focus on fast yet accurate identification. It's that far because the YOLOv8 comes in as prowess, from empirical validation with an impressive 98.9% on mAP50 and 91.7% on mAP50- 95, surpassing v7 and v9. This achievement not only speaks of the model's accuracy but also reinforces the capability for real- time processing. Implications are brought forth in transportation safety and urban planning promising transformative potential. This experiment will demonstrate the potential of YOLO in the intelligent aspect of object detection for revolutionizing and enhancing several various fields. The outcome of the YOLO model is then employed to modify the traffic signal intervals based on the vehicle density of each lane in a simulation environment that has been designed using Pygame.