用证据融合提升无人机检测的可靠性，尤其在传感器失效时。
Uncertainty-Aware Multimodal Anti-UAV Detection via Evidential Fusion and Conflict-Discounted Belief Aggregation
- 通过证据理论将多模态冲突转化为不确定性，再融合判断结果。
- 在反无人机数据集上准确率达67.0%，高于单一模态（红外59.8%，可见光59.8%）。
- 适合需要高可靠性、对传感器故障敏感的实时安防场景。
反无人机感知系统在遮挡、快速运动或模态特异性失效导致传感器质量下降时仍需保持可靠。现有基于RGB与热成像的多模态系统采用确定性融合，未建模预测不确定性，也无法表达模态间分歧时的怀疑。证据深度学习（EDL）可在一次前向传播中生成每类校准后的不确定性。已有工作EDTC已应用于热成像单模态感知，但跨模态证据融合尚未研究。本文提出折扣信念融合（DBF），将跨模态冲突转化为不确定性质量，再进行流意见聚合。通过选择不确定性较低的模态确定边界框。在Anti-UAV基准测试中，多模态融合性能稳定优于任一单模态（测试准确率0.670对比红外0.604、可见光0.598），且保持实时速度（≥38 FPS）。然而，实验发现DBF与无折扣平均几乎无差异：该基准以近乎全存在为特征，缺失编码空洞导致跨模态冲突极低，使折扣步骤无效。融合后不确定性良好校准（ECE 0.057），但定位失败检测能力弱于空间方差（AUROC 0.626 vs. 0.739）。这一空结果揭示其结构性本质：基准中普遍存在的存在性与缺失编码的真空共同抑制了跨模态冲突，限定了冲突感知融合的实际收益范围。
原文摘要 · Abstract (English)
Anti-UAV perception systems must remain reliable when sensor streams degrade under occlusion, fast motion, or modality-specific failure. Existing multimodal anti-UAV systems fuse RGB and thermal streams deterministically, without modeling predictive uncertainty, and cannot express doubt when streams disagree. Evidential Deep Learning (EDL) produces calibrated per-class uncertainty in a single forward pass. EDTC already exploits this for thermal-only perception, yet cross-modal evidential fusion remains unaddressed. This paper extends EDTC to multimodal RGB-Thermal perception via Discounted Belief Fusion (DBF), which converts inter-modal conflict into uncertainty mass before aggregating stream opinions. Bounding boxes are resolved by selecting the lower-uncertainty modality. On the Anti-UAV benchmark, multimodal fusion consistently outperforms either single stream (test Acc 0.670 vs. 0.604 IR, 0.598 RGB) at real-time speed (at least 38 FPS). However, DBF is empirically indistinguishable from undiscounted averaging: near-zero inter-modal conflict on this presence-dominated benchmark leaves the discounting step inert. The fused uncertainty is well-calibrated (ECE 0.057) yet expectedly a weaker localization failure detector than spatial variance (AUROC 0.626 vs. 0.739). The null result is structural: the benchmark's near-universal presence and vacuous miss-encoding jointly suppress inter-modal conflict, a diagnosis that delimits where conflict-aware fusion provides measurable benefit.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。