Global-Scale Self-Supervised Spatiotemporal Learning for NDVI Time-Series Reconstruction
作者: Ang Li, Menghui Jiang, Xiaobin Guan, Dong Chu, Huanfeng Shen
分类: cs.CV
发布日期: 2026-08-03
💡 一句话要点
提出GloSSR框架以解决NDVI时间序列重建问题
🎯 匹配领域: 支柱八:物理动画 (Physics-based Animation)
关键词: NDVI重建 自监督学习 时空学习 深度学习 遥感数据
📋 核心要点
- 现有方法在重建受云污染和噪声干扰的NDVI时间序列时面临数据获取困难和模型泛化能力不足的挑战。
- 本文提出GloSSR框架,通过人工降解清晰NDVI数据生成自监督训练对,并采用双向Transformer与ConvLSTM网络进行时空学习。
- 实验结果表明,GloSSR在人工降级像素重建中优于对比方法,并能准确捕捉植被动态和关键的物候状态。
📝 摘要(中文)
准确高效地重建受云污染和噪声干扰的NDVI时间序列在遥感领域仍然是一个挑战。深度学习为建模复杂的时空依赖关系提供了有希望的解决方案,但由于获取相同时空位置的清晰和退化NDVI数据对的困难,其应用受到限制。为了解决这一问题,本文提出了GloSSR,一个全球尺度自监督时空框架用于NDVI重建。该框架通过人工降解相对干净的NDVI观测,构造监督信号,生成自监督训练对,进一步引入了一个端到端的时空学习网络,通过双向Transformer与ConvLSTM架构共同捕捉长程时间依赖和短期时空相关性。大量评估表明,该框架在人工和真实场景中均表现出色。
🔬 方法详解
问题定义:本文旨在解决云污染和噪声干扰下的NDVI时间序列重建问题。现有方法往往依赖于清晰和退化数据对的获取,导致模型训练受限。
核心思路:GloSSR框架通过人工降解清晰的NDVI观测,生成自监督训练对,模拟真实世界的降解情况,从而克服数据获取的困难。
技术框架:该框架包括一个端到端的时空学习网络,结合双向Transformer和ConvLSTM架构,能够捕捉长程时间依赖和短期时空相关性。同时,设计了一个基于时间通道注意力的重建模块和时空先验约束,以优化重建效果。
关键创新:GloSSR的主要创新在于自监督学习策略,通过人工降解生成训练对,避免了对清晰数据的依赖,且引入的时空先验约束有效保留了细微结构和长期物候趋势。
关键设计:在网络结构上,采用双向Transformer与ConvLSTM的结合,设计了时间通道注意力模块以增强信息特征,损失函数则考虑了重建误差与时空先验约束的综合优化。
🖼️ 关键图片
📊 实验亮点
在实验中,GloSSR在人工降级像素重建任务中表现优异, consistently outperforming 其他对比方法,显示出在NDVI时间序列重建中的有效性。基于真实观测的时间序列分析表明,该框架能够准确捕捉植被动态和关键物候状态,验证了其在长期植被趋势分析中的应用潜力。
🎯 应用场景
该研究具有广泛的应用潜力,特别是在大规模环境监测、农业监测和生态研究等领域。通过准确重建NDVI时间序列,能够更好地理解植被动态和生态变化,为决策提供数据支持。未来,该框架可扩展至其他遥感数据集,进一步提升其应用价值。
📄 摘要(原文)
Accurate and efficient reconstruction of cloud-contaminated and noise-corrupted NDVI time series remains a challenge in remote sensing. Deep learning provides a promising solution for modeling complex spatiotemporal dependencies; however, its application is often limited by the difficulty of obtaining paired clear-sky and degraded NDVI data for identical spatiotemporal locations. To address this issue, we propose GloSSR, a Global-scale Self-supervised Spatiotemporal framework for NDVI Reconstruction. The framework constructs supervisory signals by artificially degrading relatively clean NDVI observations with realistic cloud contamination patterns, producing self-supervised training pairs that closely mimic real-world degradation. It further introduces an end-to-end spatiotemporal learning network that jointly captures long-range temporal dependencies and short-term spatiotemporal correlation through a bidirectional Transformer with a ConvLSTM architecture. A temporal-channel attention-based reconstruction module is incorporated to enhance informative features, while a spatiotemporal prior constraint is designed to preserve both fine-scale structures and long-term phenological trends during optimization. Extensive evaluations on MODIS NDVI data demonstrate the effectiveness of the proposed framework across both artificial and real-world scenarios. In artificial degraded-pixel reconstruction experiments, GloSSR consistently outperforms the comparison methods. Time-series analyses based on real observations further demonstrate that the proposed framework can accurately characterize vegetation dynamics and capture the key phenological states. Long-term vegetation trend analysis and the transferability analysis to AVHRR data validate the scalability of the framework and illustrate its broad applicability for large-scale environmental monitoring.