Cloud Detection Algorithm Research Articles

Retrieval of cloud microphysical properties from Himawari-8/AHI infrared channels and its application in surface shortwave downward radiation estimation in the sun glint region

Satellite remote sensing of cloud property retrieval and shortwave downward radiation (SWDR) estimation is essential for global radiation budget and climate change studies. Sun glint areas remain a challenge for the existing cloud and SWDR algorithms based on the visible channel since surface specular reflection has a significant impact on satellite retrieval. In this study, a set of algorithms for cloud detection and cloud microphysical parameter estimation were developed using infrared multichannel data from the new generation geostationary satellite Himawari-8 based on the random forest method. The results indicated that the cloud retrieval algorithm exhibited better performance in the sun glint areas where the official Himawari-8 products (cloud detection and cloud optical thickness) were overestimated. We developed a new SWDR estimation algorithm combining the radiative transfer model and machine learning techniques by considering the cloud properties from the cloud retrieval algorithm. The results indicated that the SWDR and cloud radiative forcing derived by the new algorithm were more consistent with those of the well-known radiation products Cloud and the Earth's Radiant Energy System than those estimated using the official-based cloud product, with decreases in the root mean square error of approximately 22% and 41%, respectively. The new algorithms effectively addressed sun glint contamination by providing more data coverage and exhibiting stable performance on a spatiotemporal scale.

An improved method for retrieving aerosol optical depth over Ebinur Lake Basin from Gaofen-1

Lake Ebinur is the largest saltwater lake in Xinjiang. In recent years, the exposed dry lake bottom surface has become a remarkable source of salt and dust owing to the desiccation of the lake. Given the importance of the lake basin to human livelihoods, comprehensive monitoring of basin behavior is required. Remote sensing offers numerous novel approaches to monitoring the spatial and temporal distribution of aerosols. During this study, high spatial resolution aerosol optical depth (AOD) inversion at Lake Ebinur was still in the gap stage. The wide-field-of-view (WFV) data from four cameras on the GaoFen-1 (GF-1) satellite, launched by China in April 2013, offers high spatial and temporal resolution and excellent potential for AOD estimation. Owing to the shortwave infrared band (SWIR) deficiency and the complexity of surface reflectance, the coefficient of variation (CV) was proposed as a measure of surface reflectance to develop a surface reflectance database for land cover characteristics in the Ebinur Lake Basin. The significant improvements of the algorithm include (1) the design of an accurate and universal cloud detection algorithm through the blue and red bands. (2) The CV was used to assess the robustness of surface reflectance under long time series. After dividing the surface reflectance image pixels with low robustness, a reasonable value was assigned according to the surface reflectance variation characteristics of the image pixels. (3) Aerosol models were determined from the long-term measurements of the CIMEL solar photometer (CE318), and the validation of ground-based observations showed good correlation that WFV/GF-1 AOD was comparable to CE318 AOD (R = 0.705; root mean square error (RMSE) = 0.098; mean absolute error (MAE) = 0.078), and the smaller MAE and RMSE confirmed the feasibility of inversion of AOD by this method.

An integrated approach to reconstructing snow cover under clouds and cloud shadows on Sentinel-2 Time-Series images in a mountainous area

Snow is an important type of surface cover and an important freshwater resource in the arid regions of northwest China, and optical satellites are an important means for monitoring variation in snow cover. However, the radiation values measured by satellite sensors are affected by cloud cover and cloud shadows; thus, the reconstruction of snow cover under clouds and cloud shadows becomes a bottleneck problem especially in mountains areas. Based on the digital elevation model (DEM), in this paper, an integrated method of snow cover reconstruction on Sentinel-2 images is created by combining the improved SNOWL (snow line) algorithm considering unstable snow cover areas, Fmask cloud detection algorithms, and the Sen2Cor clear sky snow cover detection tool with a geospatial processing platform, GEE (Google Earth Engine). Considering the Babao River Basin as the study area, 474 Sentinel-2 scenes (10 m) were selected to obtain the snow cover spatiotemporal variations of 158 days from November 2016 to March 2021. Compared with the high-resolution GF-2 satellite images, the integrated method can extract snow under clouds and cloud shadows well, and the improved SNOWL algorithm improves the overall accuracy from 66.05 % to 84.26 %. The experimental results show that approximately half of the Babao River Basin is unstable snow cover, and areas above 3977 m are covered with snow for at least half a year.

CIFNet: context information fusion network for cloud and cloud shadow detection in optical remote sensing imagery

Accurately and automatically detecting cloud and cloud shadow is one of the key steps in the analysis of optical remote sensing imagery. Currently, most cloud and cloud shadow detection methods are prone to false detection, and some clouds and cloud shadows may be missing from the detection results. Due to the lack of contextual information extraction and fusion capabilities, the accuracy of these cloud detection algorithms cannot be guaranteed. This article proposes a deep learning-based convolutional neural network context information fusion network (named CIFNet) to solve the problem of cloud and cloud shadow detection on Gaofen-1 wide field of view optical remote sensing imageries. First, we preprocess the dataset and make the corresponding labels. Next, we design a Resblock-cloud to capture global and local features and prevent the network from degrading. Then, we utilize a global context fusion module to fuse different levels of global context information by dense skip-connections. Finally, a multiscale context fusion module is designed to extract multiscale contextual relations between cloud and cloud shadow. Experimental results show that the proposed CIFNet method can obtain better cloud and cloud shadow boundaries in complex situations and therefore outperforms the competing methods.

Image Compression Network Structure Based on Multiscale Region of Interest Attention Network

In this study, we proposed a region of interest (ROI) compression algorithm under the deep learning self-encoder framework to improve the reconstruction performance of the image and reduce the distortion of the ROI. First, we adopted a remote sensing image cloud detection algorithm for detecting important targets in images, that is, separating the remote sensing background from important regions in remote sensing images and then determining the target regions because most traditional ROI-based image compression algorithms utilize the manual labeling of the ROI to achieve region separation in images. We designed a multiscale ROI self-coding network from coarse to fine with a hierarchical super priority layer to synthesize images to reduce the spatial redundancy more effectively, thus greatly improving the distortion rate performance of image compression. By using a spatial attention mechanism for the ROI in the image compression network, we achieved better compression performance.

Satellite Image Compression Guided by Regions of Interest

Small satellites empower different applications for an affordable price. By dealing with a limited capacity for using instruments with high power consumption or high data-rate requirements, small satellite missions usually focus on specific monitoring and observation tasks. Considering that multispectral and hyperspectral sensors generate a significant amount of data subjected to communication channel impairments, bandwidth constraint is an important challenge in data transmission. That issue is addressed mainly by source and channel coding techniques aiming at an effective transmission. This paper targets a significant further bandwidth reduction by proposing an on-the-fly analysis on the satellite to decide which information is effectively useful before coding and transmitting. The images are tiled and classified using a set of detection algorithms after defining the least relevant content for general remote sensing applications. The methodology makes use of the red-band, green-band, blue-band, and near-infrared-band measurements to perform the classification of the content by managing a cloud detection algorithm, a change detection algorithm, and a vessel detection algorithm. Experiments for a set of typical scenarios of summer and winter days in Stockholm, Sweden, were conducted, and the results show that non-important content can be identified and discarded without compromising the predefined useful information for water and dry-land regions. For the evaluated images, only 22.3% of the information would need to be transmitted to the ground station to ensure the acquisition of all the important content, which illustrates the merits of the proposed method. Furthermore, the embedded platform's constraints regarding processing time were analyzed by running the detection algorithms on Unibap's iX10-100 space cloud platform.

A two-branch cloud detection algorithm based on the fusion of a feature enhancement module and Gaussian mixture model.

Accurate cloud detection is an important step to improve the utilization rate of remote sensing (RS). However, existing cloud detection algorithms have difficulty in identifying edge clouds and broken clouds. Therefore, based on the channel data of the Himawari-8 satellite, this work proposes a method that combines the feature enhancement module with the Gaussian mixture model (GMM). First, statistical analysis using the probability density functions (PDFs) of spectral data from clouds and underlying surface pixels was conducted, selecting cluster features suitable for daytime and nighttime. Then, in this work, the Laplacian operator is introduced to enhance the spectral features of cloud edges and broken clouds. Additionally, enhanced spectral features are input into the debugged GMM model for cloud detection. Validation against visual interpretation shows promising consistency, with the proposed algorithm outperforming other methods such as RF, KNN and GMM in accuracy metrics, demonstrating its potential for high-precision cloud detection in RS images.

A Cloud Detection Algorithm for Early Morning Observations From the FY-3E Satellite

A Cloud Detection Algorithm for Early Morning Observations From the FY-3E Satellite

A Priori Land Surface Reflectance Synergized With Multiscale Features Convolution Neural Network for MODIS Imagery Cloud Detection

Moderate Resolution Imaging Spectrometer (MODIS) images are widely used in land, ocean, and atmospheric monitoring, due to their wide spectral coverage, high temporal resolution, and convenient data acquisition. Accurate cloud detection is critical to the fine processing and application of MODIS images. Owing to spatial resolution limitations and the influence of mixed pixels, most MODIS cloud detection algorithms struggle to effectively recognize of clouds and ground objects. Here, we propose a novel cloud detection method based on land surface reflectance (LSR) and a multiscale feature convolutional neural network to achieve high-precision cloud detection, particularly for thin clouds and clouds over bright surface. A monthly surface reflectance dataset was constructed by MODIS products (MOD09A1) and employed to provide background information for cloud detection. Difference-based samples were obtained using surface reflectance as well MODIS images of different phases based on difference operations. The multiscale feature network (MFCD-Net) using an atrous spatial pyramid pooling (ASPP) and a channel and spatial attention module (CSAM) integrated low-level spatial features and high-level semantic information to capture multiscale features and generate a high-precision cloud mask. For cloud detection experiments and quantitative analysis, 61 MODIS images acquired at different times on various underlying surface types were used. Cloud detection results were compared to those of UNet, Deeplabv3+, UNet++, PSPNet, and top of atmosphere-based (MFCD-TOA) methods. The proposed method performed well, with the highest overall accuracy (96.55%), precision (92.13%), and recall (88.90%). It improved cloud detection accuracy in various scenarios, reducing thin cloud omission and bright surface misidentification.

Cloud-Detection Algorithm for Images Obtained Using the Visual and Infrared Multispectral Imager

Cloud-Detection Algorithm for Images Obtained Using the Visual and Infrared Multispectral Imager

CloudSEN12, a global dataset for semantic understanding of cloud and cloud shadow in Sentinel-2

Accurately characterizing clouds and their shadows is a long-standing problem in the Earth Observation community. Recent works showcase the necessity to improve cloud detection methods for imagery acquired by the Sentinel-2 satellites. However, the lack of consensus and transparency in existing reference datasets hampers the benchmarking of current cloud detection methods. Exploiting the analysis-ready data offered by the Copernicus program, we created CloudSEN12, a new multi-temporal global dataset to foster research in cloud and cloud shadow detection. CloudSEN12 has 49,400 image patches, including (1) Sentinel-2 level-1C and level-2A multi-spectral data, (2) Sentinel-1 synthetic aperture radar data, (3) auxiliary remote sensing products, (4) different hand-crafted annotations to label the presence of thick and thin clouds and cloud shadows, and (5) the results from eight state-of-the-art cloud detection algorithms. At present, CloudSEN12 exceeds all previous efforts in terms of annotation richness, scene variability, geographic distribution, metadata complexity, quality control, and number of samples.

Joint multiscale cloud detection algorithm for ground-based lidar.

A ground-based lidar is a powerful tool for studying the vertical structure and optical properties of clouds. A layer detection algorithm is important to determine the presence and spatial position of clouds from vast lidar signals. However, current detection algorithms for ground-based lidar still involve substantial missing and false detections for tenuous layers and layer edges. Here, a joint multiscale cloud layer detection algorithm is proposed. The algorithm can effectively capture the tenuous layers and layer edges by using joint multiscale detection methods based on a trend function and the Bernoulli distribution assumption. Results show that the proposed algorithm detects 10.45% more cloud layers than the official cloud product of Micro Pulse Lidar Network (MPLNET) does. Specifically, 7.93% and 12.57% more cloud layers are detected at daytime and nighttime, respectively. The evaluation based on depolarization properties proves that the additional cloud layers detected by the joint multiscale algorithm are reliable. These additional detected clouds have important implications for cloud climatology and climate change research. The new algorithm remarkably enhances the cloud detection capability of ground-based lidar and potentially be widely used by the community.

A CO2-independent cloud mask from Infrared Atmospheric Sounding Interferometer (IASI) radiances for climate applications

Abstract. With more than 15 years of continuous and consistent measurements, the Infrared Atmospheric Sounding Interferometer (IASI) radiance dataset is becoming a reference climate data record. To be exploited to its full potential, it requires a cloud filter that is accurate, unbiased over the full IASI life span and strict enough to be used in satellite data retrieval schemes. Here, we present a new cloud detection algorithm which combines (1) a high sensitivity, (2) a good consistency over the whole IASI time series and between the different copies of the instrument flying on board the suite of Metop satellites, and (3) simplicity in its parametrization. The method is based on a supervised neural network (NN) and relies, as input parameters, on the IASI radiance measurements only. The robustness of the cloud mask over time is ensured in particular by avoiding the IASI channels that are influenced by CO2, N2O, CH4, CFC-11 and CFC-12 absorption lines and those corresponding to the ν2 H2O absorption band. As a reference dataset for the training, version 6.5 of the operational IASI Level 2 (L2) cloud product is used. We provide different illustrations of the NN cloud product, including comparisons with other existing products. We find very good agreement overall with version 6.5 of the operational IASI L2 with an identical mean annual cloud amount and a pixel-by-pixel correspondence of about 87 %. The comparison with the other cloud products shows a good correspondence in the main cloud regimes but with sometimes large differences in the mean cloud amount (up to 10 %) due to the specificities of each of the different products. We also show the good capability of the NN product to differentiate clouds from dust plumes.

Cloud Detection of Gaofen-2 Multi-Spectral Imagery Based on the Modified Radiation Transmittance Map

For optical remote sensing images with high spatial resolution and low spectral number, the complexity of ground objects poses great challenges to cloud detection algorithms, such as the differentiation of clouds from objects with similar features as clouds and the identification of thin clouds. In this paper, a novel cloud detection method is proposed for Gaofen-2 remote sensing imagery. The radiation transmittance is estimated based on the dark channel prior, and the overestimated radiation transmittance is corrected using spectral features. A three-step post-processing strategy is adopted to eliminate misidentification introduced by the highlighted surfaces based on object geometric, textural, and boundary features. In the experiments, Gaofen-2 multispectral images with different cloud categories and cloud thicknesses are involved to evaluate the performance of the proposed method. The results show that the proposed method can obtain an average cloud detection accuracy of 0.9573 on six different clouds. The proposed algorithm can also effectively detect both thick and thin clouds with an average accuracy of more than 0.9517. The advantages of the method for thin cloud detection are further demonstrated by comparison with existing algorithms.

Data-Free Area Detection and Evaluation for Marine Satellite Data Products

The uncertainty verification of satellite ocean color products and the bias analysis of multiple data are both indispensable in the evaluation of ocean color products. Incidentally, ocean color products often have missing information that causes the methods mentioned above to be difficult to evaluate these data effectively. We propose an analysis and evaluation method based on data-free area. The objective of this study is to evaluate the quality of ocean color products with respect to information integrity and continuity. First, we use an improved Spectral Angle Mapper, also called ISAM. It can automatically obtain the optimal threshold value for each class of objects. Then, based on ISAM, we perform spectral information mining on first-level Yellow Sea and Bohai Sea data obtained from the Geostationary Ocean Color Imager (GOCI), Moderate Resolution Imaging Spectroradiometer (MODIS) and Ocean and Land Color Instrument (OLCI). In this manner, quantitative results of information related to data-free areas of ocean data products are obtained. The findings indicate that the product data of OLCI are optimal with respect to both completeness and continuity. GOCI and MODIS have striking similarities in their quantitative or visualization results for both evaluation metrics. Moreover, a concomitant phenomenon of ocean-covered objects is apparent in the data-free area with temporal and spatial distribution characteristics. The two characteristics are subsequently explored for further analysis. The evaluation method adopted in this study can help to enrich the content of ocean color product evaluation, facilitate the research of cloud detection algorithms and further understand the composition of the data-free regional information of marine data products. The method proposed in this study has a wide application value.

Cloud Detection of Remote Sensing Image Based on Multi-Scale Data and Dual-Channel Attention Mechanism

Cloud detection is one of the critical tasks in remote sensing image preprocessing. Remote sensing images usually contain multi-dimensional information, which is not utilized entirely in existing deep learning methods. This paper proposes a novel cloud detection algorithm based on multi-scale input and dual-channel attention mechanisms. Firstly, we remodeled the original data to a multi-scale layout in terms of channels and bands. Then, we introduced the dual-channel attention mechanism into the existing semantic segmentation network, to focus on both band information and angle information based on the reconstructed multi-scale data. Finally, a multi-scale fusion strategy was introduced to combine band information and angle information simultaneously. Overall, in the experiments undertaken in this paper, the proposed method achieved a pixel accuracy of 92.66% and a category pixel accuracy of 92.51%. For cloud detection, the proposed method achieved a recall of 97.76% and an F1 of 95.06%. The intersection over union (IoU) of the proposed method was 89.63%. Both in terms of quantitative results and visual effects, the deep learning model we propose is superior to the existing semantic segmentation methods.

Evaluating the potential of H8/AHI geostationary observations for monitoring vegetation phenology over different ecosystem types in northern China

• Different strategies for compositing daily EVI2 from 10-min AHI data were assessed. • Mean value compositing using data around solar noon minimized anisotropic effects. • Composited AHI daily EVI2 time series have 5.3 times more observations than MODIS. • Phenology dates were successfully retrieved from composited AHI EVI2 time series. • AHI can reduce uncertainty in phenology monitoring with more valid data and less gaps. Sensors onboard the new generation of geostationary (GEO) satellites launched over past few years have much improved radiometric and temporal resolutions. As one of the latest GEO sensors, the Japanese Himawari-8 Advanced Himawari Imager (hereafter H8/AHI) has temporal resolution of 10 minutes and spectral bands similar to MODIS. Observations from H8/AHI therefore have the potential to improve the precision of vegetation phenology monitoring by significantly shortening the compositing period. In this study, we evaluated methods for compositing daily 2-band Enhanced Vegetation Index (EVI2) from 10-minutes H8/AHI observations at six study sites in northern China encompassing forest, cropland, and grassland. After screening the cloudy observations by a physics-based cloud detection algorithm, key phenological metrics were retrieved from the composited AHI EVI2. H8/AHI 10-min raw EVI2 showed significant diurnal variations that are mainly associated with sun angle variations. By calculating the daily average EVI2 values in different time intervals, we concluded that the mean value using a fixed time window (10:30–13:30) centered around local solar noon effectively reduced the anisotropy effect from the diurnal sun-angle variations, leading to stable EVI2 time series suitable for extracting phenological metrics. On average, the amount of the composited H8/AHI EVI2 time-series data was 5.3 times more than the Terra/Aqua combined MODIS EVI2 product per year and the average size (width) of the gaps in H8/AHI EVI2 time series was 75% smaller than that of the MODIS. Our results demonstrated the promising capability of the new generation GEO satellite for generating time series vegetation index with fewer cloud-induced gaps and higher temporal resolution, hence improving the monitoring of global vegetation phenology and ecosystem responses to climate change.

Hybrid Cloud Detection Algorithm Based on Intelligent Scene Recognition

Abstract All-sky images derived from ground-based imaging equipment have become an important means of recognizing and quantifying cloud information. Accurate cloud detection is a prerequisite for obtaining important cloud information from an all-sky image. Existing cloud segmentation algorithms can achieve high accuracy. However, for different scenes, such as completely cloudy with obscured sun and partly cloudy with unobscured sun, the use of specific algorithms can further improve segmentation. In this study, a hybrid cloud detection algorithm based on intelligent scene recognition (HCD-ISR) is proposed. It uses suitable cloud segmentation algorithms for images in different scenes recognized by ISR, so as to utilize the various algorithms to their full potential. First, we developed an ISR method to automatically classify the all-sky images into three scenes. In scene A, the sky is completely clear; in scene B, the sky is partly cloudy with unobscured sun; and in scene C, the sun is completely obscured by clouds. The experimental results show that the ISR method can correctly identify 93% of the images. The most suitable cloud detection algorithm was selected for each scene based on the relevant features of the images in that scene. A fixed thresholding (FT) method was used for the images in scene C. For the most complicated scene, that is, scene B, the clear-sky background difference (CSBD) method was used to identify cloud pixels based on a clear-sky library (CSL). The images in the CSL were automatically filtered by ISR. Compared to FT, adaptive thresholding (AT), and CSBD methods, the proposed HCD-ISR method has the highest accuracy (95.62%). The quantitative evaluation and visualization results show that the proposed HCD-ISR algorithm makes full use of the advantages of different cloud detection methods, and is more flexible and robust.

ACCURACY ASSESSMENT OF CLOUD MASK DETECTION ALGORITHMS FOR CBERS-4 WFI IMAGERY

Abstract. Clouds limit the potential use of optical images. Proper clouds and cloud shadows detection are necessary steps for optical image applications. Few algorithms are flexible in detecting clouds in images with a limited number of bands, such as the Wide-Field Imager (WFI) sensor on board the China-Brazil Earth Resources Satellite (CBERS-4), which has four spectral bands (blue, green, red, and near-infrared). Therefore, this work aims to assess the accuracy of two cloud detection algorithms: CMASK and ATSA, and evaluate the influence of the ATSA configuration parameters. We selected four regions in Brazil for our analysis. In all cases, ATSA presented overall accuracy (OA) superior to CMASK. While the ATSA obtained OA greater than 0.91 for all analyzes, the OA from CMASK did not exceed 0.84. CMASK presented commission errors for the No Clear class (combination of Cloud and Cloud shadow) and inclusion errors for the Clear class close to zero. However, many errors of omission of clouds misclassified as the Clear class was observed. The ATSA algorithm presented a better balance between inclusion errors and omission errors. Our results can be used as guidance for choosing a cloud mask algorithm for the CBERS-4 WFI images and for analysis considering the images from WFI on board CBERS-4A and Amazonia-1, as they have similar characteristics.

Bayesian Cloud Detection over Land for Climate Data Records

Cloud detection is a necessary step in the generation of land surface temperature (LST) climate data records (CDRs) and affects data quality and uncertainty. We present here a sensor-independent Bayesian cloud detection algorithm and show that it is suitable for use in the production of LST CDRs. We evaluate the performance of the cloud detection with reference to two manually masked datasets for the Advanced Along-Track Scanning Radiometer (AATSR) and find a 7.9% increase in the hit rate and 4.9% decrease in the false alarm rate when compared to the operational cloud mask. We then apply the algorithm to four instruments aboard polar-orbiting satellites, which together can produce a global, 25-year LST CDR: the second Along-Track Scanning Radiometer (ATSR-2), AATSR, the Moderate Resolution Spectroradiometer (MODIS Terra) and the Sea and Land Surface Temperature Radiometer (SLSTR-A). The Bayesian cloud detection hit rate is assessed with respect to in situ ceilometer measurements for periods of overlap between sensors. The consistency of the hit rate is assessed between sensors, with mean differences in the cloud hit rate of 4.5% for ATSR-2 vs. AATSR, 4.9% for AATSR vs. MODIS, and 2.5% for MODIS vs. SLSTR-A. This is important because consistent cloud detection performance is needed for the observational stability of a CDR. The application of a sensor-independent cloud detection scheme in the production of CDRs is thus shown to be a viable approach to achieving LST observational stability over time.