Reconstruction of Cloud-free Sentinel-2 Image Time-series Using an Extended Spatiotemporal Image Fusion Approach

Abstract

Time-series for medium spatial resolution satellite imagery are a valuable resource for environmental assessment and monitoring at regional and local scales. Sentinel-2 satellites from the European Space Agency (ESA) feature a multispectral instrument (MSI) with 13 spectral bands and spatial resolutions from 10 m to 60 m, offering a revisit range from 5 days at the equator to a daily approach of the poles. Since their launch, the Sentinel-2 MSI image time-series from satellites have been used widely in various environmental studies. However, the values of Sentinel-2 image time-series have not been fully realized and their usage is impeded by cloud contamination on images, especially in cloudy regions. To increase cloud-free image availability and usage of the time-series, this study attempted to reconstruct a Sentinel-2 cloud-free image time-series using an extended spatiotemporal image fusion approach. First, a spatiotemporal image fusion model was applied to predict synthetic Sentinel-2 images when clear-sky images were not available. Second, the cloudy and cloud shadow pixels of the cloud contaminated images were identified based on analysis of the differences of the synthetic and observation image pairs. Third, the cloudy and cloud shadow pixels were replaced by the corresponding pixels of its synthetic image. Lastly, the pixels from the synthetic image were radiometrically calibrated to the observation image via a normalization process. With these processes, we can reconstruct a full length cloud-free Sentinel-2 MSI image time-series to maximize the values of observation information by keeping observed cloud-free pixels and calibrating the synthetized images by using the observed cloud-free pixels as references for better quality.

The Application of Remote Sensing Technologies for Enhancing Agroecosystems Performance

“…A regular time resolution affects data availability in various stages of the crop development process. However, the data will not be helpful when clouds are present in the area to be examined [15]. Conversely, drones fly several hundred meters above the ground.…”

Reconstruction of Sentinel-2 Image Time Series Using Google Earth Engine

“…Normalized Difference Vegetation Index (NDVI) is one of the most widely used vegetation indexes [2,3], it is simple to calculate and can be used for crop yield estimation [4], leaf area index (LAI) estimation [5] and other studies. Surface reflectance and NDVI time series have been widely used in land cover classification, environmental change, crop monitoring and other fields, and have become important indicators to describe global climate change and its impact on terrestrial ecosystems [3,[6][7][8][9]. There are many related geoscience studies mainly based on the surface reflectance and NDVI time series data of Advanced Very High Resolution Radiometer (AVHRR), Moderate Resolution Imaging Spectroradiometer (MODIS), Landsat and Sentinel-2 images.…”

“…The Sentinel-2 mission provides a combination of two satellites, Sentinel-2a and Sentinel-2b, equipped with the Multi Spectral Instrument (MSI) sensor with 13 bands from the visible and the near infrared (NIR) to short wave infrared spectral range (SWIR) at different spatial resolutions from 10 to 60 m, including the red edge band that are beneficial for vegetation physiology research [10]. Sentinel-2 images have become one of the most valuable Earth observation resources in remote sensing studies and applications [7,[11][12][13]. However, clouds are common in remote sensing images which cause the retrieved surface parameters incomplete, and residual clouds or undetected cloud shadows can still be observed in Sentinel-2 surface reflectance products.…”

“…Many methods have been developed to reconstruct the surface reflectance or NDVI time series. Based on several studies [7,14,15], methods to reconstruct surface reflectance time series can be divided into four types: (1) temporal-based methods [16,17]; (2) spectralbased methods [18,19]; (3) hybrid methods [20,21]; (4) multi-source fusion [7,[22][23][24]. In comparison, methods to reconstruct NDVI time series can be divided into four categories:…”

“…Sentinel-2 data is similar to Landsat data but with higher spatio-temporal resolution. Some researchers [7,39,40] used the multi-source fusion method to reconstruct the Sentinel-2 image time series, among them, Zhou et al [7] used a spatiotemporal image fusion approach, the prediction smooth reflectance fusion model (PSRFM) [24], to reconstruct the cloud-free Sentinel-2 image time series, and achieved good results; Xiong et al [40] proposed a deep learning-based fusion algorithm, enhanced residual dense network (ERDN), that can simultaneously fuse Landsat-7,8 and Sentinel-2 surface reflectance to generate dense Sentinel-2 image time series. However, spatiotemporal or deep learning-based fusion approaches faces the problems of spatial resolution inconsistency and spectral band difference, and the latter method requires high computational power and is time-consuming.…”