Vehicular mobility monitoring using remote sensing and deep learning on a UAV-based mobile computing platform

Basit Öğe Kaydı
dc.contributor.authorBakirci, Murat
dc.date.accessioned2025-03-17T12:27:02Z
dc.date.available2025-03-17T12:27:02Z
dc.date.issued2025
dc.departmentTarsus Üniversitesi
dc.description.abstractConventional methods of transportation security, such as fixed security camera surveillance, offer limited coverage and constrained viewing angles. Moreover, the full potential of unmanned systems, widely utilized in various fields, remains underexploited in transportation security. This study integrates Intelligent Transportation Systems (ITS) with Unmanned Aerial Vehicles (UAVs) to address these limitations and propose a comprehensive approach to urban mobility monitoring and control. By enhancing both the software and hardware of a test UAV, the system is transformed into a robust mobile computing platform capable of real-time vehicle detection and traffic monitoring. The UAV's upgraded communication infrastructure enables rapid data transmission to control stations, facilitating timely decision-making. Various size variants of YOLOv8 are tested for their suitability for real-time applications, with the YOLOv8n variant proving particularly effective, achieving faster detection times and a precision rate of 0.706 in vehicle detection tasks on the Jetson Nano platform. YOLOv8n was also compared against various variants of the state-of-the-art algorithms YOLOv9, YOLOv10, and YOLO11. While it did not achieve the highest precision, the difference was minor, and it demonstrated the best balance between precision and recall. The system reliably tracks specific vehicles, providing accurate real-time vehicle trajectories to the control station. Additionally, using the UAV's hovering mode, the study successfully maps traffic density at critical transportation nodes, ensuring consistent monitoring and control.
dc.identifier.doi10.1016/j.measurement.2024.116579
dc.identifier.issn0263-2241
dc.identifier.issn1873-412X
dc.identifier.scopus2-s2.0-85213083249
dc.identifier.scopusqualityQ3
dc.identifier.urihttps://doi.org/10.1016/j.measurement.2024.116579
dc.identifier.urihttps://hdl.handle.net/20.500.13099/2021
dc.identifier.volume244
dc.identifier.wosWOS:001402542900001
dc.identifier.wosqualityQ1
dc.indekslendigikaynakWeb of Science
dc.indekslendigikaynakScopus
dc.institutionauthorBakirci, Murat
dc.language.isoen
dc.publisherElsevier Sci Ltd
dc.relation.ispartofMeasurement
dc.relation.publicationcategoryMakale - Uluslararası Hakemli Dergi - Kurum Öğretim Elemanı
dc.rightsinfo:eu-repo/semantics/closedAccess
dc.snmzKA_WOS_20250316
dc.subjectUrban mobility
dc.subjectVehicle detection
dc.subjectIntelligent transportation systems
dc.subjectUAV
dc.subjectMobile computing platform
dc.subjectYOLOv8n
dc.subjectYOLO11
dc.titleVehicular mobility monitoring using remote sensing and deep learning on a UAV-based mobile computing platform
dc.typeArticle

Dosyalar

Koleksiyon

WoS İndeksli Yayınlar Koleksiyonu
Scopus İndeksli Yayınlar Koleksiyonu