Spatiotemporal Fusion Network for Land Surface Temperature Based on a Conditional Variational Autoencoder

High spatiotemporal resolution land surface temperature (LST) data are essential for dynamic monitoring and prediction in climate change research. Due to the limitations of remote sensing instruments, the current platforms have difficulty in achieving a compromise between high spatial and temporal r...

Celý popis

Uloženo v:
Podrobná bibliografie
Vydáno v:IEEE transactions on geoscience and remote sensing Ročník 60; s. 1 - 13
Hlavní autoři: Chen, Yuncheng, Yang, Yingbao, Pan, Xin, Meng, Xiangjin, Hu, Jia
Médium: Journal Article
Jazyk:angličtina
Vydáno: New York IEEE 2022
The Institute of Electrical and Electronics Engineers, Inc. (IEEE)
Témata:
Biological system modeling
Climate change
Climate change research
Climate prediction
Coders
Conditional variational autoencoder (CVAE)
convolutional neural network (CNN)
Data models
Deep learning
Heterogeneity
Image reconstruction
Instruments
Land surface temperature
land surface temperature (LST)
Noise reduction
Outliers (statistics)
pretraining
Process parameters
Remote sensing
River basins
Root-mean-square errors
Spatial resolution
spatiotemporal fusion
Spatiotemporal phenomena
Surface temperature
Temperature sensors
Training
Weighting
Work platforms
ISSN:0196-2892, 1558-0644
On-line přístup:Získat plný text
Tagy: Přidat tag
Žádné tagy, Buďte první, kdo vytvoří štítek k tomuto záznamu!
Popis
Shrnutí:High spatiotemporal resolution land surface temperature (LST) data are essential for dynamic monitoring and prediction in climate change research. Due to the limitations of remote sensing instruments, the current platforms have difficulty in achieving a compromise between high spatial and temporal resolutions for LST products. In this study, we propose a spatiotemporal fusion network for LST based on a conditional variational autoencoder (CVAE-LSTFM). First, an improved network is designed based on the CVAE by reconstructing an encoder and a decoder. To generate fine LST images based on dense time series, a variational inference model is formulated to integrate coarse and fine LST image pairs in variational autoencoded latent space. In addition, a new compound loss function for the proposed training method is designed to reduce the effects of noise and outliers. Then, a pretraining mechanism is adopted to optimize the network training process, and the parameters can be transferred to the training network of the CVAE to accelerate network convergence. Finally, a novel weighting strategy that considers spatiotemporal variations in LST (LST consistency weighting) is employed to solve the spatiotemporal heterogeneity problem caused by the rapid changes in LST. The method is quantitatively tested and evaluated in the Heihe River Basin using FY-4A LST and MODIS LST from September 2019. Compared with two traditional models and two deep learning-based models, CVAE-LSTFM yields lower root-mean-square error (RMSE) (average < 1.26 K) and learning perceptual image patch similarity (LPIPS, average < 0.13) values and higher structural similarity (SSIM, average <inline-formula> <tex-math notation="LaTeX">>0.96 </tex-math></inline-formula>). In practice, CVAE-LSTFM can generate high spatiotemporal resolution LST values (hourly LST with a 1-km spatial resolution) with high accuracy, quality, and robustness.
Bibliografie:ObjectType-Article-1
SourceType-Scholarly Journals-1
ObjectType-Feature-2
content type line 14
ISSN:0196-2892
1558-0644
DOI:10.1109/TGRS.2022.3183114