Pi-Vision: Leveraging YOLO V8 for Real-Time Person Detection and Recognition on Raspberry Pi

Surveillance, crowd management, and person tracking are major problems. This research work describes the process of counting and identifying people in real-world scenarios, as well as tracking the movement of vehicles in traffic and the number of cars entering and leaving parking areas. This pioneers the innovative integration of YOLOv8 with the Raspberry Pi 4B, revolutionizing people's counting and identification systems. By optimizing YOLOv8's architecture and training methodologies, the approach achieves real-time person detection and recognition. Large dataset curation, model refinement, and evaluation techniques ensure accurate individual counting and identification, signifying its potential for security surveillance and crowd management. The research highlights YOLOv8's adaptability and efficacy in diverse settings, particularly when integrated with Raspberry Pi 4B. The simulation results depict that the suggested model achieved an accuracy of 90.29%.