UAV-SGDFNet: structured guided dual-level feature fusion for low-altitude anti-UAV detection

Abstract

In complex low-altitude environments, illegally operated UAVs are typically small and can easily be confused with background clutter, leading to reduced detection stability. To address this challenge, we propose UAV-SGDFNet, an efficient small UAV detector. The model enhances robustness by jointly optimizing feature representation and multi-scale fusion. In the feature extraction phase, we introduce a spatial-channel collaborative attention mechanism to strengthen discriminative features for the target and suppress background interference. In the feature fusion phase, we design a structurally guided dual-level fusion module that enables structured cross-level interaction through the cooperation of a guidance path and a fidelity path, mitigating semantic dilution and improving the localization of small targets. Experimental results on a complex-scene dataset show that the proposed method outperforms several mainstream detectors, achieving an mAP@0.5 of 0.748 and increasing Recall to 0.692, while maintaining high inference efficiency, indicating a favorable accuracy-efficiency trade-off. Ablation studies and visual analyses further confirm the effectiveness of the proposed components for stable small-target detection and indicate its potential for real-time deployment.

Keywords

Anti-UAV detection / Small object detection / Attention mechanism / Real-time detection

References

[1] Khawaja W., Ezuma M., Semkin V., Erden F., Ozdemir O., Guvenc I.. A survey on detection, classification, and tracking of UAVs using radar and communications systems. IEEE Commun. Surv. Tutor., 2026, 28: 3272-3310

[2] Ezuma M., Anjinappa C.K., Semkin V., Guvenc I.. Comparative analysis of radar-cross-section-based UAV recognition techniques. IEEE Sens. J., 2022, 22(1817932-17949

[3] Ezuma M., Anjinappa C.K., Funderburk M., Guvenc I.. Radar cross section based statistical recognition of UAVs at microwave frequencies. IEEE Trans. Aerosp. Electron. Syst., 2021, 58(1): 27-46

[4] Rudys S., Laučys A., Ragulis P., Aleksiejūnas R., Stankevičius K., Kinka M., et al.. Hostile UAV detection and neutralization using a UAV system. Drones, 2022, 6(9 250

[5] Inani K.N., Sangwan K.S., et al.. Machine learning based framework for drone detection and identification using RF signals. 2023 Int. Conf. Innovative Trends Inf. Technol. (ICITIIT), 2023IEEE, 1-8

[6] X. Tu, C. Zhang, H. Zhuang, S. Liu, R. Li, Fast drone detection with optimized feature capture and modeling algorithms. IEEE Access (2024)

[7] T. Sangam, I.R. Dave, W. Sultani, M. Shah, Transvisdrone: spatio-temporal transformer for vision-based drone-to-drone detection in aerial videos. IEEE Access (2022). arXiv:2210.08423

[8] Rebbapragada S.V.C., Panda P., Balasubramanian V.N.. C2FDrone: coarse-to-fine drone-to-drone detection using vision transformer networks. 2024 IEEE Int. Conf. Robot. Autom. (ICRA), 2024IEEE, 6627-6633

[9] Zhang F., Zhou S., Wang Y., Wang X., Hou Y.. Label assignment matters: a Gaussian assignment strategy for tiny object detection. IEEE Trans. Geosci. Remote Sens., 2024, 62: 1-12

[10] Zhang J., Lin Y., Zhou X., Shi P., Zhu X., Zeng D.. Precision in pursuit: a multi-consistency joint approach for infrared anti-UAV tracking. Vis. Comput., 2025, 41(4): 2187-2202

[11] Huang B., Dou Z., Chen J., Li J., Shen N., Wang Y., et al.. Searching region-free and template-free Siamese network for tracking drones in TIR videos. IEEE Trans. Geosci. Remote Sens., 2023, 62: 1-15

[12] Wisniewski M., Rana Z.A., Petrunin I., Holt A., Harman S.. Towards fully autonomous drone tracking by a reinforcement learning agent controlling a pan–tilt–zoom camera. Drones, 2024, 8(6 235

[13] P. Bhattacharya, P. Nowak, Drone detection and tracking with YOLO and a rule-based method

[14] Wang H., Dang Z., Cui M., Shi H., Qu Y., Ye H., et al.. ADG-YOLO: a lightweight and efficient framework for real-time UAV target detection and ranging. Drones, 2025, 9(10 707

[15] Zhou S., Yang L., Liu H., Zhou C., Liu J., Wang Y., et al.. Improved YOLO for long range detection of small drones. Sci. Rep., 2025, 15(1 12280

[16] Laroca R., Dos Santos M., Menotti D.. Improving small drone detection through multi-scale processing and data augmentation. 2025 International Joint Conference on Neural Networks (IJCNN), 2025IEEE, 1-8

[17] Kurmashev I., Semenyuk V., Lupidi A., Alyoshin D., Kurmasheva L., Cantelli-Forti A.. Study of the optimal YOLO visual detector model for enhancing UAV detection and classification in optoelectronic channels of sensor fusion systems. Drones, 2025, 9(11 732

[18] Si Y., Xu H., Zhu X., Zhang W., Dong Y., Chen Y., et al.. SCSA: exploring the synergistic effects between spatial and channel attention. Neurocomputing, 2025, 634 129866

[19] Zhang Y., Bao W., Yang Y., Wan W., Xiao Q., Zou X.. HAFNet: hierarchical attention fusion network for infrared small target detection. IEEE Trans. Geosci. Remote Sens., 2025, 63: 1-16

[20] Zhao Y., Lv W., Xu S., Wei J., Wang G., Dang Q., et al.. DETRs beat YOLOs on real-time object detection. Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, 202416965-16974

[21] Lv K., Xiao Z., Ren H., Li X., Lan P.. DGE-YOLO: dual-branch gathering and attention for efficient accurate UAV object detection. Appl. Sci., 2026, 16(8 4004

[22] Zhang B., Wang L., Yao Q., Li K., Tan Q.. SG-YOLO: a multispectral small-object detector for UAV imagery based on YOLO. Remote Sens., 2026, 18(7 1003

[23] H. Zhang, K. Liu, Z. Gan, G.N. Zhu, UAV-DETR: efficient end-to-end object detection for unmanned aerial vehicle imagery. arXiv preprint (2025). arXiv:2501.01855

[24] D. Lee, S. Yun, Y. Ro, Emulating: self-attention with convolution for efficient image super-resolution. arXiv preprint (2025). arXiv:2503.06671

[25] Woo S., Park J., Lee J.Y., Kweon I.S.. CBAM: convolutional block attention module. Proc. Eur. Conf. Comput. Vis. (ECCV), 20183-19

[26] Y. Liu, Z. Shao, N. Hoffmann, Global: attention mechanism: retain information to enhance channel-spatial interactions. CoRR (2021). arXiv:2112.05561

[27] Wang Q., Wu B., Zhu P., Li P., Zuo W., Hu Q.. ECA-Net: efficient channel attention for deep convolutional neural networks. Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR), 202011531-11539

[28] Y. Wang, J. Zhang, M. Kan, S. Shan, X. Chen, Self-supervised equivariant attention mechanism for weakly supervised semantic segmentation. CoRR (2020). arXiv:2004.04581

[29] Ouyang D., He S., Zhang G., Luo M., Guo H., Zhan J., et al.. Efficient multi-scale attention module with cross-spatial learning. Proceedings of the IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP), 20231-5

Funding

Guangzhou Municipal Science and Technology Program key projects(Grant 62233013)

Rights & permissions

The Author(s)