A Unified Framework for Vehicle Detection, Tracking, and Counting Across Ground and Aerial Views Using Knowledge Distillation with YOLOv10-S

Abstract

Highlights: What are the main findings? The proposed unified and modular framework effectively integrates vehicle detection, multi-object tracking, and trajectory-based counting into a consistent end-to-end pipeline that works across both ground and aerial surveillance scenarios. Knowledge distillation strengthens the lightweight YOLOv10-S detector without architectural modification, improving temporal stability and enhancing overall system reliability across diverse viewpoints. What are the implications of the main findings? The study underscores the importance of evaluating traffic monitoring from a system-level perspective, where coordinated detection and tracking directly influence counting accuracy and robustness. The proposed framework provides a practical and scalable foundation for intelligent transportation applications, offering cross-domain adaptability and real-time feasibility for real-world traffic monitoring. Accurate and reliable vehicle detection, tracking, and counting across different surveillance platforms are fundamental requirements for developing smart Traffic Management Systems (TMS) and promoting sustainable urban mobility. Recent advances in both ground-level surveillance and remote sensing using deep learning have opened new opportunities for extracting detailed vehicular information from high-resolution aerial and surveillance video data. Our research reported here aims to present a unified, real-time vehicle analysis framework that integrates lightweight deep learning–based detection, robust multi-object tracking, and trajectory-driven counting within a single modular pipeline. The proposed framework employs a “You Only Look Once” system, YOLOv10-S as the detection backbone and enhances its robustness through supervision-level knowledge distillation without introducing any architectural modifications. Temporal consistency is enforced using an observation-centric multi-object tracking algorithm (OC-SORT), enabling stable identity preservation under camera motion and dense traffic conditions. Vehicle counting is performed using a trajectory-based virtual gate strategy, reducing duplicate counts and improving counting reliability. Comprehensive experiments conducted on the UA-DETRAC and VisDrone benchmarks show that the proposed framework effectively balances detection performance, tracking robustness, counting accuracy, and real-time efficiency in both ground-based and aerial surveillance settings. Furthermore, cross-dataset evaluations under direct train–test transfer highlight the inherent challenges of domain shift while showing that knowledge distillation consistently improves robustness in detection, tracking identity consistency, and vehicle counting. Overall, this framework enables effective real-world traffic monitoring by adopting a scalable and practical system design, where reliability is prioritized over architectural complexity.