Optimized Small Object Detection Algorithm for UAV Perspective in Complex Traffic Scenarios
IEEE Access · 2026 · 14 · 89323-89336
DOI: 10.1109/access.2026.3702195 ↗Kurz gefasst
Fachartikel in der Zeitschrift IEEE Access, 2026. Verfasst von Shijun Liu und 2 weiteren Autor:innen. Die Arbeit nennt 42 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0).
Abstract (Originalsprache)
Small object detection in UAV vision systems faces difficulties when the traffic environment is complex because the detection accuracy is limited and the feature representation is not sufficient. Balancing real-time performance with robustness can be hard. This paper proposes OSOD-UAV, an optimized small object detection algorithm for complex traffic scenarios. First, a Spatial Adaptive Dynamic Convolution module is introduced into the backbone network. It can adjust the convolution weights. Second, the object distribution is discrete from UAV perspective and modeling long-range dependencies is difficult. A Bi-Directional Axial Attention module is designed to handle this. This module decouples the spatial dimensions. It enhances global contextual information interaction and can improve the model’s perception ability in occluded and dense scenes. Finally, the WIoU (Wise-IoU) loss function is used instead of the traditional CIoU. A dynamic focusing mechanism is used to optimize the bounding box regression process and the detection accuracy is expected to improve. The experiments show that the model achieves 32.8% in mAP@0.5. It is worth noting that the accuracy and robustness of small object detection in complex UAV scenes can be enhanced. Also, it provides a feasible solution for UAV visual detection in complex traffic flow environments.
Zitieren
APA
Liu, S., Zhu, H., & Li, P. (2026). Optimized Small Object Detection Algorithm for UAV Perspective in Complex Traffic Scenarios. IEEE Access, 14, 89323-89336. https://doi.org/10.1109/access.2026.3702195
Vancouver
Liu S, Zhu H, Li P. Optimized Small Object Detection Algorithm for UAV Perspective in Complex Traffic Scenarios. IEEE Access. 2026;14:89323-89336. doi:10.1109/access.2026.3702195
BibTeX
@article{Liu2026,
author = {Liu, Shijun and Zhu, Honghao and Li, Pengcheng},
title = {Optimized Small Object Detection Algorithm for UAV Perspective in Complex Traffic Scenarios},
journal = {IEEE Access},
year = {2026},
volume = {14},
pages = {89323--89336},
doi = {10.1109/access.2026.3702195}
}
EUVIMED