Cloudless Images Research Articles

Evaluation of Cloud Masking Methods using Sentinel-2 Satellite Images on Google Earth Engine: A Case Study in Vietnam

The production of cloudless images from the optical satellite are critical in Earth surface monitoring. In 2015,Sentinel-2A was successfully launched into orbit by the European Space Agency. Sentinel-2 imagery is currently theprimary source of data for Earth monitoring. There are several ways to create cloudless images from multi-temporalSentinel-2 optical satellite imagery on the Google Earth Engine (GEE) platform. These include the Fmask (Function ofmask) method, the Fmask CDI (Cloud Displacement Index) method, and the Fmask CSP (Cloud Score Plus) method. Inthis paper, the authors build a program and evaluate the cloud masking methods on the GEE platform in Song Hinhdistrict, Phu Yen province, which is situated in the South-Central Coast region of Central Vietnam. The Song Hinh districtis a suitable study area for the evaluation of cloud masking methods on optical satellite images due to its diverse andcomplex terrain, which includes numerous peaks and valleys and a variety of climatic conditions. This article illustratesthe results of three cloud masking methods on Sentinel-2 images. In contrast to the Fmask method, the Fmask CDI andFmask CSP methods provide more benefits in detecting clouds and cloud shadows, resulting in more accurate outcomes

Abandoned Farmland Extraction and Feature Analysis Based on Multi-Sensor Fused Normalized Difference Vegetation Index Time Series—A Case Study in Western Mianchi County

Farmland abandonment monitoring is one of the key aspects of land use and land cover research, as well as being an important prerequisite for ecological environmental protection and food security. A Normalized Difference Vegetation Index (NDVI) time series analysis is a common method used for farmland abandonment data extraction; however, extracting this information using high-resolution data is still difficult due to the limitations caused by cloud influence and data of low temporal resolution. To address this problem, this study used STARFM for GF-6 and Landsat 8 data fusion to enhance the continuity of high-resolution and cloudless images. A dataset was constructed by combining the phenological cycle of crops in the study area and then extracting abandoned farmland data based on an NDVI time series analysis. The overall accuracy of the results based on the NDVI time series analysis using the STARFM-fused dataset was 93.42%, which was 15.5% higher than the accuracy of the results obtained using only GF-6 data and 28.52% higher than those obtained using only Landsat data. Improvements in accuracy were also achieved when using SVM for time series analysis based on the fused dataset, indicating that the method used in this study can effectively improve the accuracy of the results. Then, we analyzed the spatial distribution pattern of abandoned farmland by extracting the results and concluded that the abandonment rate increased with the increase in the road network density and decreased with the increase in the distance to residential areas. This study can provide decision-making guidance and scientific and technological support for the monitoring of farmland abandonment and can facilitate the analysis of abandonment mechanisms in the study area, which is conducive to the sustainable development of farmland.

Weakly-supervised cloud detection and effective cloud removal for remote sensing images

We propose a one-stop automatic cloud processing scheme for RSI, including weakly supervised cloud detection and effective cloud removal. First, to avoid using massive expensive paired-labeled data, we use the idea of adversarial training to detect clouds, that is, randomly selecting some unpaired cloud and cloudless images to alternatively train a spectral-normalized Markov discriminator to obtain cloud masks. Second, a two-stage downsampling-restoring-upsampling-refining scheme is used to remove detected clouds. In order to further improve the rationality and fineness, we construct a fractional-order anisotropic filter kernel and design a convolution process to impose spatial regularization constraints on the loss construction at each stage, taking to account confidence values and structural priorities. Comprehensive experimental results show that our proposed networks are superior to the SOTA methods in terms of objective indicators and subjective performance. In addition, the performance of our one-stop approach in automatically detecting and eliminating clouds is also very satisfactory.

Mangrove monitoring revealed by MDPrePost-Net using archived Landsat imageries

Acting as a global vital entity in the coastal ecosystem, mangroves are currently facing major threats to destruction due to anthropogenic activities. Restoration and rehabilitation measures of mangroves are being carried out, including in Indonesia. To support conservation actions in many islands in Indonesia, spatio-temporal information on mangroves is required. Remote sensing with advanced techniques such as machine learning and deep learning has proven the capability to spatially observe mangroves. This study aims to monitor mangrove change in three study areas, i.e., South Sumatra, North Kalimantan, and Southeast Sulawesi, using a fully convolutional network (FCN)-based MDPrePost-Net. This method was developed originally to assess the mangrove degradation due to a major event (i.e., Hurricane irma 2017 in southwest Florida), whereas this study adopts it for an extended observation period (1989–2022 for South Sumatra, 1991–2021 for North Kalimantan, and 1990–2021 for Southeast Sulawesi) using medium-resolution Landsat imageries. We observed that the mangroves remain stable within the national parks designated by the government. Outside the national parks, mangrove conversion massively occurred even though the areas are assigned as protection forests. The classification results showed satisfactory accuracy values of more than 84%. The maps produced have an advantage in the spatial change analysis compared to the global datasets such as Global Mangrove Watch version 3.0. Our method has a limitation when cloudless images are not available. The integration with Synthetic Aperture Radar (SAR) images and a rigorous cloud removal method for the optical images may improve the results of mangrove monitoring.

Former-CR: A Transformer-Based Thick Cloud Removal Method with Optical and SAR Imagery

In the field of remote sensing, cloud and cloud shadow will result in optical remote sensing image contamination, particularly high cloud cover, which will result in the complete loss of certain ground object information. The presence of thick cloud severely limits the use of optical images in production and scientific research, so it is critical to conduct further research into removing the thick cloud occlusion in optical images to improve the utilization rate of optical images. The state-of-the-art cloud removal methods proposed are largely based on convolutional neural network (CNN). However, due to CNN’s inability to gather global content information, those cloud removal approaches cannot be improved further. Inspired by the transformer and multisource image fusion cloud removal method, we propose a transformer-based cloud removal method (Former-CR), which directly reconstructs cloudless images from SAR images and cloudy optical images. The transformer-based model can efficiently extract and fuse global and local context information in SAR and optical images, generating high-quality cloudless images with higher global consistency. In order to enhance the global structure, local details, and visual effect of the reconstructed image, we design a new loss function to guide the image reconstruction. A comparison with several SAR-based cloud removal methods through qualitative and quantitative experimental evaluation on the SEN12MS-CR dataset demonstrates that our proposed method is effective and superior.

Estimating fractional cover of saltmarsh vegetation species in coastal wetlands in the Yellow River Delta, China using ensemble learning model

Saltmarshes in coastal wetlands provide important ecosystem services. Satellite remote sensing has been widely used for mapping and classification of saltmarsh vegetation, however, medium-spatial-resolution satellite datasets such as Landsat-series imagery may induce mixed pixel problems over saltmarsh landscapes which are spatially heterogeneous. Sub-pixel fractional cover estimation of saltmarsh vegetation at species level are required to better understand the distribution and canopy structure of saltmarsh vegetation. In this study, we presented an approach framework for estimating and mapping the fractional cover of major saltmarsh species in the Yellow River Delta, China based on time series Landsat 8 Operational Land Imager data. To solve the problem that the coastal area is frequently covered by clouds, we adopted the recently developed virtual image-based cloud removal (VICR) algorithm to reconstruct missing image values under the cloud/cloud shadows over the time series Landsat imagery. Then, we developed an ensemble learning model (ELM), which incorporates Random Forest Regression (RFR), K-Nearest Neighbor Regression (KNNR) and Gradient Boosted Regression Tree (GBRT) based on temporal-spectral features derived from the time-series cloudless images to estimate the fractional cover of major vegetation types, i.e., Phragmites australis, Suaeda salsa and the invasive species, Spartina alterniflora. High spatial resolution imagery acquired by the Unmanned Aerial Vehicle and Gaofen-6 satellites were used for reference sample collections. The results showed that our approach successfully estimated the fractional cover of each saltmarsh species (average of R-square:0.891, RMSE: 7.48%). Through four scenarios of experiments, we found that the ELM is advantageous over each individual model. When the images during key months were absent, cloud removal for the Landsat images considerably improved the estimation accuracies. In the study area, Spartina alterniflora covers the largest area (5753.97 ha), followed by Phragmites australis with spatial extent area of 4208.4 ha and Suaeda salsa of 1984.41 ha. The average fractional cover of S. alterniflora was 58.45%, that of P. australis was 51.64% and that of S.salsa was 51.64%.

Study Model for Information Reconstruction on Cloud Contaminated Area for Single Multispectral Remote Sensing Sentinel-2 Imagery using Generative Adversarial Network

Cloud and cloud shadow cause information loss in optical remote sensing analysis. South East Asia, especially Vietnam, Sentinel-2 imagery has short re-visit cycle and observations tend to be contaminated with cloud and cloud shadow. Traditional cloud removal methods require close date multi-temporal data to avoid seasonal land cover changes. In this study, a method of integrating Deep Convolutional Neural Networks (DCNN) and Generative Adversarial Network (GAN) was proposed. This machine learning model estimates the information loss over cloud contaminated areas on a single Sentinel-2 image. The results show that for images with cloud cover rate under 25%, our model can reconstruct cloudless images with PSNR (25 – 40 dB) and SSIM (0.86 – 0.93) compared to real clear images. On the other hand, with cloud cover rate up to 40%, the model performance will be affected heavily by the distribution of cloud and cloud shadow areas. By investigating DCNN and GAN, our method has proven to be an effective tool to remove cloudy images with low and medium rates, which enriches the clear optical remote sensing data sources for environment monitoring.

Fusion of GF and MODIS Data for Regional-Scale Grassland Community Classification with EVI2 Time-Series and Phenological Features

Satellite-borne multispectral data are suitable for regional-scale grassland community classification owing to comprehensive coverage. However, the spectral similarity of different communities makes it challenging to distinguish them based on a single multispectral data. To address this issue, we proposed a support vector machine (SVM)–based method integrating multispectral data, two-band enhanced vegetation index (EVI2) time-series, and phenological features extracted from Chinese GaoFen (GF)-1/6 satellite with (16 m) spatial and (2 d) temporal resolution. To obtain cloud-free images, the Enhanced Spatial and Temporal Adaptive Reflectance Fusion Model (ESTARFM) algorithm was employed in this study. By using the algorithm on the coarse cloudless images at the same or similar time as the fine images with cloud cover, the cloudless fine images were obtained, and the cloudless EVI2 time-series and phenological features were generated. The developed method was applied to identify grassland communities in Ordos, China. The results show that the Caragana pumila Pojark, Caragana davazamcii Sanchir and Salix schwerinii E. L. Wolf grassland, the Potaninia mongolica Maxim, Ammopiptanthus mongolicus S. H. Cheng and Tetraena mongolica Maxim grassland, the Caryopteris mongholica Bunge and Artemisia ordosica Krasch grassland, the Calligonum mongolicum Turcz grassland, and the Stipa breviflora Griseb and Stipa bungeana Trin grassland are distinguished with an overall accuracy of 87.25%. The results highlight that, compared to multispectral data only, the addition of EVI2 time-series and phenological features improves the classification accuracy by 9.63% and 14.7%, respectively, and even by 27.36% when these two features are combined together, and indicate the advantage of the fine images in this study, compared to 500 m moderate-resolution imaging spectroradiometer (MODIS) data, which are commonly used for grassland classification at regional scale, while using 16 m GF data suggests a 23.96% increase in classification accuracy with the same extracted features. This study indicates that the proposed method is suitable for regional-scale grassland community classification.

Dealing With Clouds and Seasonal Changes for Center Pivot Irrigation Systems Detection Using Instance Segmentation in Sentinel-2 Time Series

The automatic detection of Center Pivot Irrigation Systems (CPIS) is fundamental for establishing public policies, especially in countries with a growth perspective in this technology, like Brazil. Previous studies to detect CPIS using deep learning used single-date optical images, containing limitations due to seasonal changes and cloud cover. Therefore, this research aimed to detect CPIS using Sentinel-2 multitemporal images (containing six dates) and instance segmentation, considering seasonal variations and different proportions of cloudy images, generalizing the models to detect CPIS in diverse situations. We used a novel augmentation strategy, in which, for each iteration, six images were randomly selected from the time series (from a total of 11 dates) in random order. We evaluated the Mask-RCNN model with the ResNext-101 backbone considering the COCO metrics on six testing sets with different ratios of cloudless ( $ ) and cloudy images ( $>75\%$ ), from six cloudless images and zero cloudy images (6:0) up to one cloudless image and five cloudy images (1:5). We found that using six cloudless images provided the best metrics [80% average precision (AP), 93% AP with a 0.5 intersection over union threshold (AP50)]. However, results were similar (74% AP, 88% AP50) even in extreme scenarios with abundant cloud presence (1:5 ratio). Our method provides a more adaptive and automatic way to map CPIS from time series, significantly reducing interference such as cloud cover, atmospheric effects, shadow, missing data, and lack of contrast with the surrounding vegetation.

Time-Series Analysis of Surface-Water Quality in Xiong\u2019an New Area, 2016\u20132019

The Xiong'an New Area was established by the Chinese government on 1 April 2017 as a new district in the Hebei Province, and soon after its establishment, investigations on the waterbodies of the district began to initiate governance. In this study, data from 105 Sentinel-2 cloudless images for the period April 2016–September 2019 were analysed to understand the changes in the concentrations of chlorophyll-a and water turbidity in surface waterbodies of the Xiong'an New Area. The results suggest that (1) the water quality shows an overall improving trend owing to surface-water pollution control measures by the Chinese government in the region; (2) the turbidity of the surface waterbody declined gradually particularly after the spring of 2017; (3) the concentration of chlorophyll-a shows a weak decreasing trend, with a low value in spring and a high value in summer with a four-year average of 31.3 μg/L; and (4) by comparing the results of the water quality of two ponds inside and outside the Lake Baiyangdian and the overall water quality of Xiong'an New Area, the changes in turbidity and chlorophyll-a concentration during the study period were consistent.

THE SPATIAL AND TEMPORAL AVAILABILITY DIFFERENCES OF CLOUD-FREE LANDSAT IMAGES OVER THREE GORGES RESERVOIR AREA

Abstract. Availability analysis of cloud-free optical remote sensing data is a prerequisite for remote sensing applications. In this study, spatio-temporal availability differences of cloud-free Landsat TM, ETM+, and OLI sensors images over Three Gorges Reservoir Area (TGRA) were analyzed from 1986 to 2019 based on the Google Earth Engine (GEE). The results show that: 1) in Summer, especially in August, the probabilities of obtaining Landsat images with no more than 30% cloud cover (CC) is higher. 2) the northeast of TGRA has higher probability of acquiring cloudless images than the southwest. 3) In TGRA, annual monitoring which require at least one cloud-free observation in a year largely unaffected by CC, but when considering seasonal monitoring, cloud contaminate will become a limitation, and monthly monitoring in this area is basically not feasible even if the three sensors data are combined. The results of this paper will provide important references for the research of using optical data in this area, and although the research area is relatively small, the analysis method and the program developed in this paper have no restrictions on the area.

Monitoring the Characteristics of the Bohai Sea Ice Using High-Resolution Geostationary Ocean Color Imager (GOCI) Data

Satellite remote sensing data, such as moderate resolution imaging spectroradiometers (MODIS) and advanced very high-resolution radiometers (AVHRR), are being widely used to monitor sea ice conditions and their variability in the Bohai Sea, the southernmost frozen sea in the Northern Hemisphere. Monitoring the characteristics of the Bohai Sea ice can provide crucial information for ice disaster prevention for marine transportation, oil field operation, and regional climate change studies. Although these satellite data cover the study area with fairly high spatial resolution, their typically limited cloudless images pose serious restrictions for continuous observation of short-term dynamics, such as sub-seasonal changes. In this study, high spatiotemporal resolution (500 m and eight images per day) geostationary ocean color imager (GOCI) data with a high proportion of cloud-free images were used to monitor the characteristics of the Bohai Sea ice, including area and thickness. An object-based feature extraction method and an albedo-based thickness inversion model were used for estimating sea ice area and thickness, respectively. To demonstrate the efficacy of the new dataset, a total of 68 GOCI images were selected to analyze the evolution of sea ice area and thickness during the winter of 2012–2013 with severe sea ice conditions. The extracted sea ice area was validated using Landsat Thematic Mapper (TM) data with higher spatial resolution, and the estimated sea ice thickness was found to be consistent with in situ observation results. The entire sea ice freezing–melting processes, including the key events such as the day with the maximum ice area and the first and last days of the frozen season, were better resolved by the high temporal-resolution GOCI data compared with MODIS or AVHRR data. Both characteristics were found to be closely correlated with cumulative freezing/melting degree days. Our study demonstrates the applicability of the GOCI data as an improved dataset for studying the Bohai Sea ice, particularly for purposes that require high temporal resolution data, such as sea ice disaster monitoring.

POSSIBILITY OF CLOUDLESS OPTICAL REMOTE SENSING IMAGES ACQUISITION STUDY BY USING METEOROLOGICAL SATELLITE OBSERVATIONS

Abstract. Optical remote sensing images have been widely used in feature interpretation and geo-information extraction. All the fundamental applications of optical remote sensing, are greatly influenced by cloud coverage. Generally, the availability of cloudless images depends on the meteorological conditions for a given area. In this study, the cloud total amount (CTA) products of the Fengyun (FY) satellite were introduced to explore the meteorological changes in a year over China. The cloud information of CTA products were tested by using ZY-3 satellite images firstly. CTA products from 2006 to 2017 were used to get relatively reliable results. The window period of cloudless images acquisition for different areas in China was then determined. This research provides a feasible way to get the cloudless images acquisition window by using meteorological observations.

Analysis the changes of annual for Poyang Lake wetland vegetation based on MODIS monitoring

Abstract This paper studies on Poyang Lake natural wetlands. we calculation each month's vegetation area in Poyang Lake that according to cloudless images of MODIS in 2009. And it had a quantitative analysis for the changes of annual by Poyang Lake wetland vegetation with time; space; and the water level. Poyang Lake wetland vegetation is mainly located in the delta of the river lower reaches. The area is changing with month, each year there are two peaks. Before the flood season that the vegetation area increase from spring to April and May. Entering the flood season, the area had been decreased by water level raised; after season, the area increase again with water level dropped. To the autumn and winter, the areas decreased cause of vegetation withered.

A Technique for estimating regional surface albedos using geostationary satellite data

AbstractData from the Japanese geostationary meteorological satellite are used to map surface albedo in Tasmania, Australia. Satellite data are corrected for atmospheric depletion and the narrow spectral window of the satellite sensor. The effect of non‐Lambertian reflection is discussed.A composite albedo map is presented based on eight mostly cloudless images taken at 10.00 and 13.00 local time in a period spanning the end of January to the beginning of February, 1985. The map features a distinct east west gradient in the albedo, with lowest values measured in the rain‐forest environment of the western highlands. Highest values are observed among areas of cleared farmland in the eastern midlands.