Towards UAV Detection in the Real World: A New Multispectral Dataset UAVNet-MS and a New Method

2026-05-20

AI Summary

Existing RGB-based drone detection methods suffer significant performance degradation in complex scenarios involving small-scale targets, low contrast, and background clutter, and there is a notable absence of multispectral datasets tailored for fine-grained micro-drone detection. To address these challenges, this work introduces UAVNet-MS, the first synchronized RGB–multispectral drone detection dataset comprising 15,618 aligned image pairs, and proposes MFDNet, a dual-stream network that effectively integrates material-aware spectral information through disparity correction and spatial–spectral feature fusion. Experimental results demonstrate that MFDNet outperforms the best RGB-only method by 6.2% in AP50, establishing a foundational dataset, a strong baseline model, and a public evaluation benchmark for multispectral drone monitoring.

Abstract

The proliferation of unmanned aerial vehicles (UAVs) has created urgent demand for precise UAV monitoring. Existing RGB-based systems rely on spatial cues that degrade at small scales, particularly with high inter-type similarity, target-clutter ambiguity, and low contrast. Multispectral imaging (MSI) encodes material-aware spectral signatures, yet MSI-based fine-grained small-UAV detection remains underexplored due to lack of dedicated datasets. We introduce UAVNet-MS, the first multispectral dataset for fine-grained small-UAV detection, comprising 15,618 temporally synchronized RGB-MSI data cubes (1440x1080) with bounding box annotations. The dataset features challenging small objects (93.7% <= 32^2 pixels, average 18^2 pixels, ~0.02% image area) under low contrast. We propose MFDNet, a dual-stream baseline addressing array-induced parallax and spatial-spectral fusion. Extensive evaluation under RGB-only, MSI-only, and RGB+MSI protocols against 20 detectors shows MFDNet achieves +6.2% AP50 improvement over best RGB-only methods, demonstrating spectral cues provide complementary material evidence beyond spatial cues. This work provides foundational dataset, strong baseline, and benchmark for multispectral UAV monitoring research.

Problem

• UAV detection
• multispectral imaging
• small object detection
• fine-grained recognition
• low contrast

Innovation

• multispectral imaging
• small UAV detection
• UAVNet-MS
• spatial-spectral fusion
• MFDNet