.png){kind=icon}
返回
Self-supervised accurate remote sensing atmospheric missing data reconstruction: a masked auto-encoder solution
.png){kind=icon}
PRE
AI
DOI：10.1080/01431161.2025.2518508.png)
.png){kind=icon}
原文求助.png){kind=icon}
分享.png){kind=icon}
收藏
摘要
En 中文
Remote sensing atmospheric products are crucial for monitoring and analysis air quality. However, missing data can severely affect the completeness and accuracy of datasets, limiting their validity and reliability in subsequent analyses and applications. This study introduced a framework based on masked auto-encoder structure tailored to reconstruct missing data in remote sensing atmospheric products. Leveraging a self-supervised approach, real missing pattern masks were employed to construct training datasets, enhancing the model’s ability to generalize and accurately recover missing values. Experiments on POMINO-TROPOMINO2vertical column density data show that by integrating residual multi-attention in the encoder and a multi-layered spatial attention in the decoder, the model achieved an overall testR2of 0.85 for datasets with less than 40% missing, showing its effectiveness and generalizability. This research contributes a novel perspective to atmospheric data recovery, offering a reliable tool for enhancing the completeness and accuracy of remote sensing atmospheric products.
Keyword:
Atmospheric data reconstruction
remote sensing
masked auto-encoder
real missing pattern
attention mechanism
期刊
IF:
2.6
论文数:
1.2W
被引数:
2.7W
机构
引用论文
Enhancing accuracy of air quality sensors with machine learning to augment large-scale monitoring networks通过机器学习提高空气质量传感器的准确性，以增强大规模监测网络
OAAI
OAAI分享收藏XGBoost.png)
Proceedings of the 22nd ACM SIGKDD International Conference on Knowledge Discovery and Data Mining
OAAI
Proceedings of the 22nd ACM SIGKDD International Conference on Knowledge Discovery and Data MiningIF0
OAAI分享收藏Superpixel-Oriented Thick Cloud Removal Method for Multitemporal Remote Sensing Images面向超像素的多时相遥感影像厚云去除方法
PREAI
分享收藏Large-scale MODIS AOD products recovery: Spatial-temporal hybrid fusion considering aerosol variation mitigation大规模MODIS AOD产品回收: 考虑气溶胶变化缓解的时空混合融合
PREAI
分享收藏
查看更多内容