时空卫星图像降尺度的迁移编码与自回归生成模型
Spatiotemporal Satellite Image Downscaling with Transfer Encoders and Autoregressive Generative Models
AI总结：
本文提出基于迁移学习和自回归生成模型的时空卫星图像降尺度方法，通过预训练U-Net编码器和扩散模型，实现高分辨率图像重建，提升环境监测效果。
摘要：
我们提出了一种基于迁移学习的生成降尺度框架，用于从粗分辨率输入中重建高分辨率卫星图像。我们的方法结合了一个轻量级的U-Net迁移编码器和基于扩散的生成模型。首先，更简单的U-Net在长时间序列的粗分辨率数据上进行预训练，以学习时空表示；然后将其编码器冻结并转移到更大的降尺度模型中，作为物理上有意义的潜在特征。我们的应用使用NASA的MERRA-2再分析数据作为低分辨率源域（50公里），以及GEOS-5自然运行（G5NR）作为高分辨率目标域（7公里）。我们的研究区域包括亚洲的一个大区域，通过将其划分为两个子区域和四个季节，使计算变得可行。我们使用Wasserstein距离进行域相似性分析，证实了MERRA-2和G5NR之间分布变化最小，验证了冻结参数转移的安全性。在季节性区域划分中，我们的模型取得了出色的性能（R2=0.65至0.94），优于确定性U-Net、变分自编码器和先前的迁移学习基线。使用半变异函数、ACF/PACF和基于滞后RMSE/R2的出数据评估表明，预测的降尺度图像保持了物理上一致的空间变化和时间自相关性，使模型能够在G5NR记录之外实现稳定的自回归重建。这些结果表明，迁移增强的扩散模型为在有限训练期内降尺度长时间序列的粗分辨率图像提供了一种稳健且物理上一致的解决方案。这一进展对改进环境暴露评估和长期环境监测具有重要意义。
英文摘要：
We present a transfer-learning generative downscaling framework to reconstruct fine resolution satellite images from coarse scale inputs. Our approach combines a lightweight U-Net transfer encoder with a diffusion-based generative model. The simpler U-Net is first pretrained on a long time series of coarse resolution data to learn spatiotemporal representations; its encoder is then frozen and transferred to a larger downscaling model as physically meaningful latent features. Our application uses NASA's MERRA-2 reanalysis as the low resolution source domain (50 km) and the GEOS-5 Nature Run (G5NR) as the high resolution target (7 km). Our study area included a large area in Asia, which was made computationally tractable by splitting into two subregions and four seasons. We conducted domain similarity analysis using Wasserstein distances confirmed minimal distributional shift between MERRA-2 and G5NR, validating the safety of parameter frozen transfer. Across seasonal regional splits, our model achieved excellent performance (R2 = 0.65 to 0.94), outperforming comparison models including deterministic U-Nets, variational autoencoders, and prior transfer learning baselines. Out of data evaluations using semivariograms, ACF/PACF, and lag-based RMSE/R2 demonstrated that the predicted downscaled images preserved physically consistent spatial variability and temporal autocorrelation, enabling stable autoregressive reconstruction beyond the G5NR record. These results show that transfer enhanced diffusion models provide a robust and physically coherent solution for downscaling a long time series of coarse resolution images with limited training periods. This advancement has significant implications for improving environmental exposure assessment and long term environmental monitoring.
机构、总结、中文摘要均由AI分析而来，可能存在错误，请以论文正文为准。
