EUVIMEDEUVIMED entdecken
Crossref

SODA-Net: A Lightweight Small Object Detection Network for Drone-Based Optical Sensor Systems

Yong Gao, Hui Jin, Shiqing Lu, Xia Huang, Jun Ding

IEEE Sensors Journal · 2026-08 · 26 (16) · 24732-24752

DOI: 10.1109/jsen.2026.3710965 ↗

Kurz gefasst

Fachartikel in der Zeitschrift IEEE Sensors Journal, 2026-08. Verfasst von Yong Gao und 4 weiteren Autor:innen. Die Arbeit nennt 52 Literaturangaben. Gefördert von: Science and Technology Research Youth Project of Chongqing Municipal Education Commission Foundation, Natural Science Foundation of Chongqing Municipality.

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

The widespread deployment of drone-based optical sensor systems in environmental monitoring, traffic management, and security surveillance has made small object detection a critical task in aerial sensing. However, the sensor-captured objects are typically small, low-resolution, and frequently occluded, posing challenges for achieving high accuracy under resource-constrained conditions. To address these challenges, this article proposes SODA-Net, a lightweight small object detection network tailored for drone optical sensor imagery, enabling efficient real-time inference on edge devices. The proposed method integrates four key modules to enhance feature representation with low computational cost. A shallow feature enhancement (SFE) module improves early feature extraction and semantic fusion. A detail enhancement and feature fusion (DEFF) module strengthens edge and texture representation using lightweight reparameterized convolutions. An attention-guided lightweight downsampling (ALDS) module preserves sensitivity to small objects while improving feature utilization. In addition, a ghost shuffle-based spatial pyramid pooling-fast module enlarges the receptive field for global context aggregation and multiscale fusion. Experiments on the VisDrone2019 dataset show that SODA-Net improves mAP50 by 4.4% while reducing parameters by 77.3%. Further validation on TinyPerson and aerial image tiny object detection (AITOD) demonstrates strong generalization, with mAP50 gains of 5.0% and 6.8%, respectively. The results indicate its effectiveness for real-time deployment in resource-limited aerial optical sensor systems.

Förderung

  • Science and Technology Research Youth Project of Chongqing Municipal Education Commission Foundation
  • Natural Science Foundation of Chongqing Municipality

Zitieren

APA

Gao, Y., Jin, H., Lu, S., Huang, X., & Ding, J. (2026). SODA-Net: A Lightweight Small Object Detection Network for Drone-Based Optical Sensor Systems. IEEE Sensors Journal, 26(16), 24732-24752. https://doi.org/10.1109/jsen.2026.3710965

Vancouver

Gao Y, Jin H, Lu S, Huang X, Ding J. SODA-Net: A Lightweight Small Object Detection Network for Drone-Based Optical Sensor Systems. IEEE Sensors Journal. 2026;26(16):24732-24752. doi:10.1109/jsen.2026.3710965

BibTeX
@article{Gao2026,
  author = {Gao, Yong and Jin, Hui and Lu, Shiqing and Huang, Xia and Ding, Jun},
  title = {SODA-Net: A Lightweight Small Object Detection Network for Drone-Based Optical Sensor Systems},
  journal = {IEEE Sensors Journal},
  year = {2026},
  volume = {26},
  number = {16},
  pages = {24732--24752},
  doi = {10.1109/jsen.2026.3710965}
}
Kostenlos starten →