REAL-TIME AUTO START-STOP YOLO V11 SINGLE SHOT DETECTION APPROACH FOR TRACKING AND SURVEILLANCE SYSTEM

Abstract

In this paper, the automotive vehicle systems auto stop and start have profoundly transformed the detection and tracking of human presence, enhancing safety and operational efficiency by using artificial intelligence (AI). Accurate vehicle
detection advances intelligent transportation, autonomous driving, and traffic monitoring systems. This study explores the capabilities of YOLOv11, the latest innovation in the YOLO deep learning model series with integration of transformers for fast learning and decision making system, tailored explicitly for vehicle detections. Developed on the strengths of its predecessors, YOLOv11 incorporates significant architectural enhancements that improve detection speed, accuracy, and robustness in dynamic and complex traffic scenarios. This study analyzes accident prevention strategies by adopting auto start and stop methodology based on advanced object detection algorithms through YOLO v11. The proposed model examines the real-time application of model adaptability in conjunction with GPS technology, emphasizing their critical roles in bolstering vehicle safety and reducing accident occurrences. The findings demonstrate that YOLO v11 significantly surpasses traditional approaches in speed and accuracy with 99.75%. The results contribute to ongoing advancements in refining Advanced Driver Assistance Systems (ADAS) and enhancing the overall driving experience.