提升卫星测深泛化能力，用深度学习+时空优化实现跨区域精准水深预测。
From Local Training to Large-Scale Mapping: A Comparative Assessment of Machine Learning and Deep Learning for Transferable Satellite-Derived Bathymetry
- 采用连续礁块训练与平滑加权损失函数，增强模型空间连续性
- 跨区域预测误差仅2.46-2.98米，3米以内精度达0.26米
- 支持多时相数据融合，适合沿海复杂区快速部署
利用多光谱遥感影像进行卫星测深（SDB）成本低但跨区域推广困难，尤其在光学复杂的近岸环境。本文评估了机器学习与深度学习在0-20米水深范围内的可迁移性，基于哨兵-2影像，在普拉塔斯岛和大堡礁部分区域训练随机森林与四种卷积神经网络（ResNet-50、ResNet-101、EfficientNet-B4、ConvNeXt-Large），并在空间独立的区域内测试。保持训练时连续礁块结构是最重要的设计选择；引入平滑权重函数（SWF）加权的均方根误差损失，强化近表层深度预测。内区域预测中，RMSE为1.15-1.92米，深度≤3米时最低达0.26米。随机森林跨区域性能显著下降（1.53→2.99-3.78米），而深度模型更稳定（2.46-2.98米）。在公开的MagicBathyNet航空RGB基准（0-16米）上，所提模型达到0.19-0.22米的RMSE，优于U-Net基线和任务特定的Transformer，且参数更少。进一步利用多时相重复影像：训练阶段扩大数据多样性，推理阶段对多次观测结果取中值，有效降低太阳角度、大气条件、水质和潮汐变化带来的噪声。研究已发布优化架构与预训练权重，支持新区域快速迁移。
原文摘要 · Abstract (English)
Satellite-derived bathymetry (SDB) from multispectral imagery is cost-effective but scales poorly across regions, especially in optically complex coastal environments. We evaluate machine learning and deep learning for transferable SDB over the 0-20 m depth range using Sentinel-2 imagery. A Random Forest baseline and four CNNs (ResNet-50, ResNet-101, EfficientNet-B4, ConvNeXt-Large) are trained on Pratas Island and selected Great Barrier Reef regions, then evaluated on spatially independent intra- and cross-regional test areas. Preserving spatial continuity during training, by keeping contiguous reef blocks rather than random patches, is the single most impactful design choice; we further introduce a Smooth Weight Function (SWF)-weighted RMSE loss that emphasizes near-surface depths. With these choices, intra-regional RMSE ranges from 1.15 to 1.92 m over 0-20 m and is as low as 0.26 m for depths <= 3 m. Random Forest degrades sharply under cross-regional transfer (RMSE 1.53 m -> 2.99-3.78 m), while the deep models stay more robust (2.46-2.98 m). On the public MagicBathyNet aerial-RGB benchmark (0-16 m) the proposed networks reach 0.19-0.22 m RMSE, outperforming a U-Net baseline and a task-specific transformer architecture with substantially fewer parameters. We further exploit multi-temporal repeat imagery: training on it broadens diversity, and median-aggregating predictions across passes at inference reduces noise from changing sun angles, atmospheric conditions, water properties, and tides. We release optimized architectures and pretrained weights to enable scalable transfer to new sites.
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。