针对灾害影像细粒度损伤识别难题，提出新型语义分割模型。
DA-SegFormer: Damage-Aware Semantic Segmentation for Fine-Grained Disaster Assessment
- 引入类别感知采样与动态难例挖掘，提升罕见损伤特征识别
- 在RescueNet数据集上达到74.61% mIoU，关键类别提升超10%
- 适用于无人机灾后高分辨率影像分析，对应急响应有实用价值
灾后快速准确的损毁评估对高效应急响应至关重要。然而，由于图像缩放导致纹理退化及极端类别不平衡，从无人机影像中识别细粒度损伤等级（如区分轻微与严重屋顶损毁）仍具挑战。本文提出DA-SegFormer，一种面向高分辨率灾后影像优化的SegFormer改进架构。方法引入类别感知采样策略以确保稀有损伤特征充分暴露，并结合在线难例挖掘（OHEM）与Dice Loss，动态聚焦于低频类别。此外，采用保持原始分辨率的推理协议，保留真实纹理细节。在RescueNet数据集上的实验表明，DA-SegFormer取得74.61%的mIoU，较基线提升2.55%。尤其在关键损伤类别上实现双位数提升：轻微损伤+11.7%，严重损伤+21.3%。
原文摘要 · Abstract (English)
Rapid and accurate damage assessment following natural disasters is critical for effective emergency response. However, identifying fine-grained damage levels (e.g., distinguishing minor from major roof damage) in UAV imagery remains challenging due to the degradation of texture cues during resizing and extreme class imbalance. We propose DA-SegFormer, a damage-aware adaptation of the SegFormer architecture optimized for high-resolution disaster imagery. Our method introduces a Class-Aware Sampling strategy to guarantee exposure to rare damage features, and it integrates Online Hard Example Mining (OHEM) with Dice Loss to dynamically focus on underrepresented classes. In addition, we employ a resolution-preserving inference protocol that maintains native texture details. Evaluated on the RescueNet dataset, DA-SegFormer achieves 74.61\% mIoU, outperforming the baseline by 2.55\%. Notably, our improvements yield double-digit gains in critical damage classes: Minor Damage (+11.7%) and Major Damage (+21.3%).
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。