Search

Three-dimensional reconstruction based on optical satellite images has always been a research hotspot in the field of photogrammetry. In particular, the 3D reconstruction of building areas has provided great help for urban planning, change detection and emergency response. The results of 3D reconstruction of satellite images are greatly affected by the input images, and this paper proposes an improvement method for 3D reconstruction of satellite images based on the generative adversarial network (GAN) image enhancement. In this method, the perceptual loss function is used to optimize the network, so that it can output high-definition satellite images for 3D reconstruction, so as to improve the completeness and accuracy of the reconstructed 3D model. We use the public benchmark dataset of satellite images to test the feasibility and effectiveness of the proposed method. The experiments show that compared with the satellite stereo pipeline (S2P) method and the bundle adjustment (BA) method, the proposed method can automatically reconstruct high-quality 3D point clouds. [ABSTRACT FROM AUTHOR]

Reconstructing the field of currents using remote sensing methods is a traditional task of oceanology. Satellite instruments such as altimeters, Doppler radars, and optical sensors are routinely used to solve oceanological problems for waters with large dimensions (open areas of the seas and oceans). The latter work well for reservoirs with significant water temperature gradients when applying the method of processing successive satellite images with varying time delays. Less commonly, areas of intense phytoplankton blooms, which are usual for productive waters, are used as markers for image matching. Such waters can be monitored with satellite ocean color sensors that provide high spatial resolution. The goal of the present paper is to investigate the possibility of reconstructing the flow field in the Gorki Reservoir, as an example of a medium-sized eutrophic reservoir, from sequent images of two different high-resolution ocean color sensors with a short time delay between images. This paper describes the field experiment under the satellite overpasses and presents the results of applying the maximum correlation method to two satellite images for the retrieval of the current field in comparison with shipboard data. It is shown that the proposed method has prospects for development. [ABSTRACT FROM AUTHOR]

Coral islands and reefs are formed by the cementation of the remains of shallow water reef-building coral polyps and other reef dwelling organisms in tropical oceans. They can be divided into coral islands, coral sandbanks, coral reefs, and coral shoals, of which, Coral shoals are located below the depth datum and are not exposed even at low tide, and sometimes are distributed at water depths exceeding 30 m. Satellite images with wide spatial–temporal coverage have played a crucial role in coral island and reef monitoring, and remote sensing data with multiple platforms, sensors, and spatial and spectral resolutions are employed. However, the accurate detection of coral shoals remains challenging mainly due to the depth effect, that is, coral shoals, especially deeper ones, have very similar spectral characteristics to the sea in optical images. Here, an optical remote sensing detection method is proposed to rapidly and accurately detect the coral shoals using a deep belief network (DBN) from optical satellite imagery. The median filter is used to filter the DBN classification results, and the appropriate filtering window is selected according to the spatial resolution of the optical images. The proposed method demonstrated outstanding performance by validating and comparing the detection results of the Yinli Shoal. Moreover, the expected results are obtained by applying this method to other coral shoals in the Xisha Islands, including the Binmei Shoal, Beibianlang, Zhanhan Shoal, Shanhudong Shoal, and Yongnan Shoal. This detection method is expected to provide the coral shoals' information rapidly once optical satellite images are available and cloud cover and tropical cyclones are satisfactory. The further integration of the detection results of coral shoals with water depth and other information can effectively ensure the safe navigation of ships. [ABSTRACT FROM AUTHOR]

Annual outbreaks of floating Ulva prolifera blooms in the Yellow Sea have caused serious local environmental and economic problems. Rapid and effective monitoring of Ulva blooms from satellite observations with wide spatial-temporal coverage can greatly enhance disaster response efforts. Various satellite sensors and remote sensing methods have been employed for Ulva detection, yet automatic and rapid Ulva detection remains challenging mainly due to complex observation scenarios present in different satellite images, and even within a single satellite image. Here, a reliable and fully automatic method was proposed for the rapid extraction of Ulva features using the Tasseled-Cap Greenness (TCG) index from satellite top-of-atmosphere reflectance (RTOA) data. Based on the TCG characteristics of Ulva and Ulva-free targets, a local adaptive threshold (LAT) approach was utilized to automatically select a TCG threshold for moving pixel windows. When tested on HY1C/D-Coastal Zone Imager (CZI) images, the proposed method, termed the TCG-LAT method, achieved over 95% Ulva detection accuracy though cross-comparison with the TCG and VBFAH indexes with a visually determined threshold. It exhibited robust performance even against complex water backgrounds and under non-optimal observing conditions with sun glint and cloud cover. The TCG-LAT method was further applied to multiple HY1C/D-CZI images for automatic Ulva bloom monitoring in the Yellow Sea in 2023. Moreover, promising results were obtained by applying the TCG-LAT method to multiple optical satellite sensors, including GF-Wide Field View Camera (GF-WFV), HJ-Charge Coupled Device (HJ-CCD), Sentinel2B-Multispectral Imager (S2B-MSI), and the Geostationary Ocean Color Imager (GOCI-II). The TCG-LAT method is poised for integration into operational systems for disaster monitoring to enable the rapid monitoring of Ulva blooms in nearshore waters, facilitated by the availability of near-real-time satellite images. [ABSTRACT FROM AUTHOR]

Mapping benthic reefs at high resolution and accuracy is vital for the management and conservation of coral habitats. Optical remote sensing data has emerged as a valuable tool for large-scale reef mapping in the past decades, with numerous data sets and methods being utilised and developed. In this study, we present a comprehensive comparison of optical remote sensing based bathymetry and benthic mapping methods. We use different optical data including WorldView-2 stereo and Sentinel-2 imagery to map the water depths of coral reef areas in the Xisha region of the South China Sea. Bathymetry data derived from photogrammetric and linear regression methods are compared to the reprocessed Ice, Cloud and land Elevation Satellite-2 (ICESat-2) data. We find that the linear regression method (root-mean-square-error, RMSE = 0.60 m) outperforms photogrammetry (RMSE = 1.02 m), and the higher resolution WorldView-2 data yields less systematic biases than Sentinel-2 data. Considering that water depths reflect changes in temperature and light, which are critical factors influencing coral reef distribution, we propose to use satellite-derived bathymetry as a feature for coral reef classification. We demonstrate that combining topography and spectral information can improve the overall mapping accuracy, particularly for compositions characterised by sharp boundaries.

Three-dimensional reconstruction based on optical satellite images has always been a research hotspot in the field of photogrammetry. In particular, the 3D reconstruction of building areas has provided great help for urban planning, change detection and emergency response. The results of 3D reconstruction of satellite images are greatly affected by the input images, and this paper proposes an improvement method for 3D reconstruction of satellite images based on the generative adversarial network (GAN) image enhancement. In this method, the perceptual loss function is used to optimize the network, so that it can output high-definition satellite images for 3D reconstruction, so as to improve the completeness and accuracy of the reconstructed 3D model. We use the public benchmark dataset of satellite images to test the feasibility and effectiveness of the proposed method. The experiments show that compared with the satellite stereo pipeline (S2P) method and the bundle adjustment (BA) method, the proposed method can automatically reconstruct high-quality 3D point clouds.

This research aimed to predict soil’s physical and chemical properties with a state-of-the-art hybrid model based on deep learning algorithms and optical satellite images in a region in the north of Iran. As dependent data, 317 soil samples (0–30 cm) were collected in field surveying and analyzed by the soil and water research institute for their physical (clay, silt, and sand) and chemical [electrical conductivity (EC), organic carbon (OC), phosphorus (P), soil reaction (pH), and potassium (K)] properties. Based on independent data, 23 remote sensing (RS) parameters (extracted from Landsat 8 optical images), 17 topographical parameters [extracted from the digital elevation model (DEM)], and four climatic parameters (derived from the meteorological organization). Spatial prediction of physical and chemical properties was implemented using a convolutional neural network (CNN), recurrent neural network (RNN), and hybrid CNN-RNN models. The evaluation results indicated that the hybrid CNN-RNN model had higher accuracy in all soil properties, followed by the RNN and CNN models. In the hybrid CNN-RNN model, pH (0.0206), EC (0.0958 dS/m), silt (0.0996%), P (0.1078 ppm), K (0.1185 ppm), sand (0.1360%), OC (0.1361%), and clay (0.1419%) had higher prediction accuracy, as determined by the root mean-squared error (RMSE) index. The hybrid CNN-RNN model proved to be the most effective for soil property prediction in this region. This finding underscores the potential of deep learning techniques in harnessing RS data for precise soil property mapping, with implications for land management and agricultural practices.

Coral islands and reefs are formed by the cementation of the remains of shallow water reef-building coral polyps and other reef dwelling organisms in tropical oceans. They can be divided into coral islands, coral sandbanks, coral reefs, and coral shoals, of which, Coral shoals are located below the depth datum and are not exposed even at low tide, and sometimes are distributed at water depths exceeding 30 m. Satellite images with wide spatial–temporal coverage have played a crucial role in coral island and reef monitoring, and remote sensing data with multiple platforms, sensors, and spatial and spectral resolutions are employed. However, the accurate detection of coral shoals remains challenging mainly due to the depth effect, that is, coral shoals, especially deeper ones, have very similar spectral characteristics to the sea in optical images. Here, an optical remote sensing detection method is proposed to rapidly and accurately detect the coral shoals using a deep belief network (DBN) from optical satellite imagery. The median filter is used to filter the DBN classification results, and the appropriate filtering window is selected according to the spatial resolution of the optical images. The proposed method demonstrated outstanding performance by validating and comparing the detection results of the Yinli Shoal. Moreover, the expected results are obtained by applying this method to other coral shoals in the Xisha Islands, including the Binmei Shoal, Beibianlang, Zhanhan Shoal, Shanhudong Shoal, and Yongnan Shoal. This detection method is expected to provide the coral shoals’ information rapidly once optical satellite images are available and cloud cover and tropical cyclones are satisfactory. The further integration of the detection results of coral shoals with water depth and other information can effectively ensure the safe navigation of ships.

Annual outbreaks of floating Ulva prolifera blooms in the Yellow Sea have caused serious local environmental and economic problems. Rapid and effective monitoring of Ulva blooms from satellite observations with wide spatial-temporal coverage can greatly enhance disaster response efforts. Various satellite sensors and remote sensing methods have been employed for Ulva detection, yet automatic and rapid Ulva detection remains challenging mainly due to complex observation scenarios present in different satellite images, and even within a single satellite image. Here, a reliable and fully automatic method was proposed for the rapid extraction of Ulva features using the Tasseled-Cap Greenness (TCG) index from satellite top-of-atmosphere reflectance (RTOA) data. Based on the TCG characteristics of Ulva and Ulva-free targets, a local adaptive threshold (LAT) approach was utilized to automatically select a TCG threshold for moving pixel windows. When tested on HY1C/D-Coastal Zone Imager (CZI) images, the proposed method, termed the TCG-LAT method, achieved over 95% Ulva detection accuracy though cross-comparison with the TCG and VBFAH indexes with a visually determined threshold. It exhibited robust performance even against complex water backgrounds and under non-optimal observing conditions with sun glint and cloud cover. The TCG-LAT method was further applied to multiple HY1C/D-CZI images for automatic Ulva bloom monitoring in the Yellow Sea in 2023. Moreover, promising results were obtained by applying the TCG-LAT method to multiple optical satellite sensors, including GF-Wide Field View Camera (GF-WFV), HJ-Charge Coupled Device (HJ-CCD), Sentinel2B-Multispectral Imager (S2B-MSI), and the Geostationary Ocean Color Imager (GOCI-II). The TCG-LAT method is poised for integration into operational systems for disaster monitoring to enable the rapid monitoring of Ulva blooms in nearshore waters, facilitated by the availability of near-real-time satellite images.

We present a new automatic procedure for updating digital topographic data from multi-source satellite imagery, which consists in the production of Digital Surface Models (DSMs) from high resolution optical satellite images, followed by a context-aware fusion that exploits the complementary characteristics of the multi-source DSMs. The fused DSM minimizes blunders and artifacts due to occlusions (e.g., the presence of clouds, snow or ash plumes) in the source images, resulting in improved accuracy and quality versus those that are not merged. The procedure has been tested to produce the 2021 digital topography of Mt Etna, whose summit area is constantly changing and shows the new peak of 3347 m on the north rim of the South East Crater. We also employ the 2021 DSM to measure the volcanic deposits emplaced in the last five years, finding about 120 million cubic meters, with a yearly average volume of about 24 million cubic meters in agreement with the large eruptive rates registered at Mt Etna since the nineteen seventies. The flexibility and modularity of the presented procedure make it easily exportable to other environmental contexts, allowing for a fast and frequent reconstruction of topographic surfaces even in extreme environments. [ABSTRACT FROM AUTHOR]

Timely mapping of flooded areas is critical to several emergency management tasks including response and recovery activities. In fact, flood crisis maps embed key information for an effective response to the natural disaster by delineating its spatial extent and impact. Crisis mapping is usually carried out by leveraging data provided by satellite or airborne optical and radar sensors. However, the processing of these kinds of data demands experienced visual interpretation in order to achieve reliable results. Furthermore, the availability of in situ observations is crucial for the production and validation of crisis maps. In this context, a frontier challenge consists in the use of Volunteered Geographic Information (VGI) as a complementary in situ data source. This paper proposes a procedure for flood mapping that integrates VGI and optical satellite imagery while requiring limited user intervention. The procedure relies on the classification of multispectral images by exploiting VGI for the semi-automatic selection of training samples. The workflow has been tested with photographs and videos shared on social media (Twitter, Flickr, and YouTube) during two flood events and classification consistency with reference products shows promising results (with Overall Accuracy ranging from 87% to 93%). Considering the limitations of social media-sourced photos, the use of QField is proposed as a dedicated application to collect metadata needed for the image classification. The research results show that the integration of high-quality VGI data and semi-automatic data processing can be beneficial for crisis map production and validation, supporting crisis management with up-to-date maps. [ABSTRACT FROM AUTHOR]

Sea surface currents are of significant importance in various scientific and maritime applications. There are several measurement techniques available to study surface currents, however, they have limitations in spatial coverage and resolution. This study presents a proof-of-concept for a new measurement principle that relies on the difference between a ship's speed relative to water and land. The approach involves estimating the ship speed vector relative to water from optical satellite imagery of Kelvin wakes. This ship speed vector is subtracted from the ship speed over ground, which is determined from Automatic Identification System (AIS) data, to estimate the surface current. A case study in the Strait of Gibraltar was performed using two months of Sentinel-2 imagery, which yielded 81 visible Kelvin wakes over 25 images. Surface currents were estimated in directions parallel and perpendicular to the ship's sailing line for each Kelvin wake. The estimated currents were validated with respect to surface currents derived from High-Frequency Radars (HFRs) and modelled currents from the Copernicus Marine Environmental Monitoring Service (CMEMS). The uncertainty in the two surface current components was estimated using triple collocation. After removing 12 data points with large ship course variability, standard deviations of 0.14 and 0.16 m s−1 were estimated for the surface currents along and across the sailing line, respectively. Despite limitations in measurement frequency due to satellite revisit times, cloud cover and Kelvin wake visibility, this new method can provide accurate estimates of sea surface currents in regions with high vessel density. [Display omitted] • New method to estimate sea surface currents. • Estimated currents show excellent agreement with High-Frequency radar observations. • Well suited to coastal regions with dense shipping traffic. [ABSTRACT FROM AUTHOR]

Peatland water table depth (WTD) and wetness have widely been monitored with optical and synthetic aperture radar (SAR) remote sensing but there is a lack of studies that have used multi-sensor data, i.e., combination of optical and SAR data. We assessed how well WTD can be monitored with remote sensing data, whether multi-sensor approach boosts explanatory capacity and whether there are differences in regression performance between data and peatland types. Our data consisted of continuous multiannual WTD data from altogether 50 restored and undrained Finnish peatlands, and optical (Landsat 5–8, Sentinel-2) and Sentinel-1 C-band SAR data processed in Google Earth Engine. We calculated random forest regressions with dependent variable being WTD and independent variables consisting of 21 optical and 10 SAR metrics. The average regression performance was moderate in multi-sensor models (R2 43.1%, nRMSE 19.8%), almost as high in optical models (R2 42.4%, nRMSE 19.9%) but considerably lower in C-band SAR models (R2 21.8%, nRMSE 23.4%) trained separately for each site. When the models included data from several sites but were trained separately for six habitat type and management option combinations, the average R2 was 40.6% for the multi-sensor models, 36.6% for optical models and 33.7% for C-band SAR models. There was considerable site-specific variation in the model performance (R2 −3.3–88.8% in the multi-sensor models ran separately for each site) and whether multi-sensor, optical or C-band SAR model performed best. The average regression performance was higher for undrained than for restored peatlands, and higher for open and sparsely treed than for densely treed peatlands. The most important variables included SWIR-based optical metrics and VV SAR backscatter. Our results suggest that optical data works usually better than does C-band SAR data in peatland WTD monitoring and multi-sensor approach increases explanatory capacity moderately little.

Recent landslide detection studies have focused on pixel-based deep learning (DL) approaches. In contrast, intuitive annotation of landslides from satellite imagery is based on distinct features rather than individual pixels. This study examines the feasibility of the integration framework of a DL model with rule-based object-based image analysis (OBIA) to detect landslides. First, we designed a ResU-Net model and then trained and tested it in the Sentinel-2 imagery. Then we developed a simple rule-based OBIA with only four rulesets, applying it first to the original image dataset and then to the same dataset plus the resulting ResU-Net heatmap. The value of each pixel in the heatmap refers to the probability that the pixel belongs to either landslide or non-landslide classes. Thus, we evaluate three scenarios: ResU-Net, OBIA, and ResU-Net-OBIA. The landslide detection maps from three different classification scenarios were compared against a manual landslide inventory map using thematic accuracy assessment metrics: precision, recall, and f1-score. Our experiments in the testing area showed that the proposed integration framework yields f1-score values 8 and 22 percentage points higher than those of the ResU-Net and OBIA approaches, respectively. [ABSTRACT FROM AUTHOR]

Timely mapping of flooded areas is critical to several emergency management tasks including response and recovery activities. In fact, flood crisis maps embed key information for an effective response to the natural disaster by delineating its spatial extent and impact. Crisis mapping is usually carried out by leveraging data provided by satellite or airborne optical and radar sensors. However, the processing of these kinds of data demands experienced visual interpretation in order to achieve reliable results. Furthermore, the availability of in situ observations is crucial for the production and validation of crisis maps. In this context, a frontier challenge consists in the use of Volunteered Geographic Information (VGI) as a complementary in situ data source. This paper proposes a procedure for flood mapping that integrates VGI and optical satellite imagery while requiring limited user intervention. The procedure relies on the classification of multispectral images by exploiting VGI for the semi-automatic selection of training samples. The workflow has been tested with photographs and videos shared on social media (Twitter, Flickr, and YouTube) during two flood events and classification consistency with reference products shows promising results (with Overall Accuracy ranging from 87% to 93%). Considering the limitations of social media-sourced photos, the use of QField is proposed as a dedicated application to collect metadata needed for the image classification. The research results show that the integration of high-quality VGI data and semi-automatic data processing can be beneficial for crisis map production and validation, supporting crisis management with up-to-date maps.

We present a new automatic procedure for updating digital topographic data from multi-source satellite imagery, which consists in the production of Digital Surface Models (DSMs) from high resolution optical satellite images, followed by a context-aware fusion that exploits the complementary characteristics of the multi-source DSMs. The fused DSM minimizes blunders and artifacts due to occlusions (e.g., the presence of clouds, snow or ash plumes) in the source images, resulting in improved accuracy and quality versus those that are not merged. The procedure has been tested to produce the 2021 digital topography of Mt Etna, whose summit area is constantly changing and shows the new peak of 3347 m on the north rim of the South East Crater. We also employ the 2021 DSM to measure the volcanic deposits emplaced in the last five years, finding about 120 million cubic meters, with a yearly average volume of about 24 million cubic meters in agreement with the large eruptive rates registered at Mt Etna since the nineteen seventies. The flexibility and modularity of the presented procedure make it easily exportable to other environmental contexts, allowing for a fast and frequent reconstruction of topographic surfaces even in extreme environments.

Ship detection using optical satellite imagery is of great significance in many applications such as traffic surveillance, pollution monitoring, etc. So far, a lot of ship detection methods have been developed for images covering open sea, offshore area and harbors. Compared to the ship detection in sea and offshore area, it is more difficult to detect ships in inland river due to several challenges. First of all, many ships in inland river are clustered together and hard to be separated from each other. Secondly, ships lying alongside the pier are very likely to be recognized as part of the pier. Thirdly, ships in inland river is usually smaller than those in the sea. A hierarchical method is proposed to detect the ships in inland river in this paper. The Regions of Interest (ROIs) are firstly extracted based on water–land segmentation using multi-spectral information. Then two kinds of ship candidates are extracted based on the panchromatic band. The isolated ships are detected by analyzing the shape of connected components and the clustered ships are detected by using mixtures multi-scale Deformable Part Models (DPM) and Histogram of Oriented Gradient (HOG). At last, a Back Propagation Neural Network (BPNN) is trained to classify the ship candidates using the multi-spectral bands. The experiments using Quickbird satellite images show that our approach is effective in ship detection and performs particularly well in separating the ships clustered together and staying alongside the pier. • The Regions of Interest (ROIs) are firstly extracted based on water–land segmentation using multi-spectral information. • Two kinds of ship candidates are extracted based on the panchromatic band. • Backpropagation Neural Network (BPNN) is trained and used to classify the ship candidates using the multi-spectral bands. [ABSTRACT FROM AUTHOR]

Reliable detection of clouds is a critical pre-processing step in optical satellite based remote sensing. Currently, most methods are based on classifying invidual pixels from their spectral signatures, therefore they do not incorporate the spatial patterns. This often leads to misclassifications of highly reflective surfaces, such as human made structures or snow/ice. Multi-temporal methods can be used to alleviate this problem, but these methods introduce new problems, such as the need of a cloud-free image of the scene. In this paper, we introduce the Remote Sensing Network (RS-Net), a deep learning model for detection of clouds in optical satellite imagery, based on the U-net architecture. The model is trained and evaluated using the Landsat 8 Biome and SPARCS datasets, and it shows state-of-the-art performance, especially over biomes with hardly distinguishable scenery, such as clouds over snowy and icy regions. In particular, the performance of the model that uses only the RGB bands is significantly improved, showing promising results for cloud detection with smaller satellites with limited multi-spectral capabilities. Furthermore, we show how training the RS-Net models on data from an existing cloud masking method, which are treated as noisy data, leads to increased performance compared to the original method. This is validated by using the Fmask algorithm to annotate the Landsat 8 datasets, and then use these annotations as training data for regularized RS-Net models, which then show improved performance compared to the Fmask algorithm. Finally, the classification time of a full Landsat 8 product is 18.0 ± 2.4 s for the largest RS-Net model, thereby making it suitable for production environments. • This algorithm show high performance, even when limited to RGB bands. • The processing time is low due to GPU based calculations. • Cloud detection performance over snow/ice is significantly improved. • This algorithm can be trained on the output of earlier cloud detection algorithms. [ABSTRACT FROM AUTHOR]

Highlights • Landsat data are used to estimate broadleaf encroachment in Irish forests. • A stratified estimator is used to estimate the area and CIs from the error matrix. • Total affected area of encroachment is 20,003 ha (±20%). • Results used by forest managers to revise operational forest management plans. Abstract In recent years, several critical issues have been identified concerning the performance of recently established spruce forest plantations in Ireland, in particular coniferous reforestation sites. More specifically, the reforestation of peatland areas in the midlands of Ireland have been subject to encroachment by broadleaf species, such as birch, and willow. These species regenerate naturally and compete with, and outgrow, the coniferous species (commonly Norway spruce) that were planted. In many cases, the growth of these trees is faster than the spruce, resulting in stands being completely encroached by approximately 15 years of age. Given the widespread nature of this problem coupled with the fragmented nature of the Irish forest estate, a project was established to use Earth observation data to predict the spatial distribution and species composition of the affected sites. The spatial distribution and associated area estimates of broadleaf encroachment within State owned spruce forests were assessed using multi-temporal Landsat satellite imagery. The overall accuracy of the encroachment map was 85.23% and a Kappa Index of Agreement of 0.58. The associated area of encroached or broadleaf dominated forests was 20,003 ha with a confidence interval of ±20% calculated using a sample-based estimator. [ABSTRACT FROM AUTHOR]

Highlights • An index is developed to detect building shadows from Sentinel-2 A MSI imagery. • The proposed shadow index was found insensitive to water and low albedo. • The proposed index over-preformed P algorithm, NSVDI, SI and the improved Fmask. • The index can detect contiguous building shadows cast by fine urban structures. Abstract Images from Multispectral Instrument (MSI) on Sentinel-2A are useful for studies of urban environments and their spatio-temporal changes. However, the frequent occurrence of building shadows in urban areas brings about great challenges in urban studies. Existing building shadow indices are not effective for Sentinel-2A MSI images due to that these indices do not make full use of the rich spectral information contained in Sentinel-2A. In this study, we propose a combinational shadow index (CSI) to address this challenge. In the formulation of CSI, three features, including the proposed shadow enhancement index (SEI), the normalized difference water index (NDWI) and the NIR band (B8), were combined to separate spectrally similar objects, such as building shadows, water and low albedo features at the Earth surface. The accuracy and robustness of CSI were tested by using six data sets from four cities, including Beijing, Shanghai, Guangzhou and Shenzhen in China. The performance of CSI was compared with three existing shadow indices, i.e., P algorithm, the normalized saturation-value difference index (NSVDI) and the shadow index (SI). Results show that CSI can detect building shadows with fine structures in both clear and cloudy images more effectively and worked well on large areas too. CSI can separate shadows from water and low albedo features as measured by a spectral discrimination index (SDI). Compared with the existing building shadow indices, CSI can improve the performance of building shadow detection by combining the feature information of SEI, NDWI and NIR band, and yielded satisfactory results for extracting building shadows from Sentinel-2A MSI imagery. [ABSTRACT FROM AUTHOR]

Abstract We use near simultaneous ocean color satellite imagery from NASA's Landsat 8 and ESA's Sentinel-2 missions to reconstruct surface currents along the northeastern shore of the Black Sea and study the spread of a small river plume formed in this area. Several times a year Landsat 8 and Sentinel-2 satellites both pass over the study area within a time interval of <10 min. Analysis of near simultaneous ocean color composites obtained during these periods provides an opportunity for accurate reconstruction of surface currents. It is especially efficient for detecting motion of frontal zones associated with river plumes which are visible in optical satellite imagery. Using an optical flow algorithm we reconstructed surface currents associated with the motion of a small plume formed by the Mzymta River in response to different wind forcing conditions. We show that the dynamics of the Mzymta plume are significantly different to those of large plumes. First, a near-field jet of the small plume is abruptly decelerated by friction with the subjacent ocean and does not form a recirculating bulge even under low wind forcing conditions. As a result, freshwater discharge does not accumulate near the river mouth and is transported to the far-field part of the plume. Second, under certain external forcing conditions the angle between wind direction and Ekman transport within the plume takes anomalously large values of 60–80°. As a result, onshore winds cause upstream accumulation of the river plume, while offshore/downwelling and upwelling winds result in downstream and offshore transport of freshwater discharge, respectively. Highlights • Near simultaneous Landsat 8 and Sentinel-2 images were used for motion detection. • Optical flow algorithm was applied to reconstruct spreading of a small river plume. • Dynamics of small plumes is significantly different as compared to large plumes. • Onshore winds cause upstream accumulation of a small river plume. • Other wind forcing cause downstream and offshore spreading of a small river plume. [ABSTRACT FROM AUTHOR]

The length of global coastline is about 356 thousand kilometers with various dynamic natural and anthropogenic. Although the number of studies on coastal landscape categorization has been increasing, it is still difficult to distinguish precisely them because the used methods commonly are traditional qualitative ones. With the leverage of remote sensing data and GIS tools, it helps categorize and identify a variety of features on land and water based on multi-source data. The aim of study is using different natural – social profile data obtained from ALOS, NOAA, and multi-temporal Landsat satellite images as input data of the convolutional-neural-network (CvNet) models for coastal landscape classification. Studies used 900 cut-line samples which represent coastal landscapes in Vietnam for training and optimizing CvNet models. As a result, nine coastal landscapes were identified including: deltas, alluvial, mature and young sand dunes, cliff, lagoon, tectonic, karst, and transitional landscapes. Three CvNet models using three different optimizer types classified the landscapes of other 1150 cut-lines in Vietnam with the accuracies about 98% and low loss function value. Excepting dalmatian, karst and delta coastal landscapes, five others distribute heterogeneous along the coasts in Vietnam. Therefore, the evaluation of additional natural components is necessary and CvNet model have ability to update new landscape types in variety of tropical nation as a step toward coastal landscape classification at both national and global scales. A coastal landscape classification system in Vietnam based on artificial intelligent models, remote sensing, GIS data. [Display omitted] • A Convolutional Neural Network to classify coastal landscapes was developed. • Different indices based on ALOS, NOAA, and Landsat data were used to classify coastal landscapes. • Five natural and anthropogenic factors were selected to identify a coastal landscape. • Nine coastal landscapes were identified in coastlines of Vietnam. • Coastal landscapes in Vietnam were separated into twenty regions with nine types. [ABSTRACT FROM AUTHOR]

On-orbit geometric calibration is a key technology to guarantee the geometric quality of high-resolution optical satellite imagery. In this paper, we present an approach for the on-orbit geometric calibration of high-resolution optical satellite imagery, focusing on two core problems: constructing an on-orbit geometric calibration model and proposing a robust calculation method. First, a rigorous geometric imaging model is constructed based on the analysis of the major error sources. Second, we construct an on-orbit geometric calibration model through performing reasonable optimizing and parameter selection of the rigorous geometric imaging model. On this basis, the calibration parameters are partially calculated with a stepwise iterative method by dividing them into two groups: external and internal calibration parameters. Furthermore, to verify the effectiveness of the proposed calibration model and methodology, on-orbit geometric calibration experiments for ZY1-02C panchromatic camera and ZY-3 three-line array camera are conducted using the reference data of the Songshan calibration test site located in the Henan Province, China. The experimental results demonstrate a certain deviation of the on-orbit calibration result from the initial design values of the calibration parameters. Therefore, on-orbit geometric calibration is necessary for optical satellite imagery. On the other hand, by choosing multiple images, which cover different areas and are acquired at different points in time to verify their geometric accuracy before and after calibration, we find that after on-orbit geometric calibration, the geometric accuracy of these images without ground control points is significantly improved. Additionally, due to the effective elimination of the internal distortion of the camera, greater geometric accuracy was achieved with less ground control points than before calibration.

Himalayan glaciers have shrunk rapidly in recent decades, but the spatial pattern of ice loss is highly variable and appears to be modulated by factors relating to individual glacier characteristics. This hinders our ability to predict their future evolution, which is vital for water resource management. The aim of this study is to assess recent glacier changes in the little-studied Annapurna Conservation Area (ACA; area: 7629 km2) in Nepal, and to explore local controls influencing their behaviour. We map changes in glacier area, surface elevation, and ice flow velocity on a large sample of glaciers (n = 162) in the ACA between 2000 and 2016. We found that total glacier area decreased by 8.5% between 2000 and 2014/15. Ice surface velocity changes between 2002 and 2016 were variable, with no clear trend of acceleration or deceleration. The mean surface elevation change for a smaller sample of glaciers (n = 72) was −0.33 ± 0.22 m a−1 between 2000 and 2013/16, which equates to a mean mass balance of −0.28 ± 0.24 m w.e. a−1. There was a trend of increasingly less negative mass balance towards the north. Glaciers that lost the most mass in the north of the ACA tended to have lower maximum elevations, bottom-heavy hypsometries, and were more likely to be avalanche-fed. However, these patterns were not apparent in glaciers in central ACA. There was no significant difference in the mean surface elevation change rate on the ablation zones of debris-covered compared with debris-free glaciers. Our work shows that glaciers in the ACA are losing area and mass at variable rates, but that the influence of local controls is complex, which introduces large uncertainties when predicting their future evolution.

In this paper, we propose a method for cloud removal in a cloud-contaminated high-resolution (HR) optical satellite image with two kinds of auxiliary images of different types: a low-resolution (LR) optical satellite composite image and a synthetic aperture radar (SAR) image. In the proposed method, we assume that cloud-contaminated and cloud-free regions have been detected accurately, then dictionary group learning (DGL) is used to establish structure correspondences between HR, LR, and SAR data from cloud-free patches, while interdictionary nonlocal joint sparse coding (INJSC) is used to estimate the universal representation coefficients of patches contaminated by clouds, and finally, cloud-contaminated HR patches can be reconstructed with their universal coefficients and the HR dictionary learned from DGL process. In this way, the missing information in the cloud-contaminated HR image can be reconstructed patch by patch. The proposed method is tested on a series of experiments on both simulated and real data. Experimental results show that both DGL and INJSC are beneficial to better reconstructing the missing information. This method is also compared against our previous work on the same topic, which adopted dictionary pair learning (DPL) and sparse coding (SC) to recover the missing information and achieved state-of-the-art performance at that time. The comparison shows that the method proposed in this paper significantly outperforms the previous one. [ABSTRACT FROM PUBLISHER]

Our knowledge of the spatio-temporal variation of river hydrological parameters is surprisingly poor. In situ gauge stations are limited in spatial and temporal coverage, and their number has been decreasing during the past decades. On the other hand, remote sensing techniques have proven their ability to measure different parameters within the Earth system. Satellite imagery, for instance, can provide variations in river area with appropriate temporal sampling. In this study, we develop an automatic algorithm for water body area monitoring based on maximum a posteriori estimation of Markov random fields. The algorithm considers pixel intensity, spatial correlation between neighboring pixels, and temporal behavior of the water body to extract accurate water masks. We solve this optimization problem using the graph cuts technique. We also measure the uncertainty associated with the determined water masks. Our method is applied over three different river reaches of Niger and Congo rivers with different hydrological characteristics. We validate the obtained river area time series by comparing with in situ river discharge and satellite altimetric water level time series. Along the Niger River, we obtain correlation coefficients of 0.85-0.96 for river reaches and 0.65 for the Congo River, which is demonstrably an improvement over other river mask retrieval algorithms. [ABSTRACT FROM AUTHOR]

Our knowledge of the spatio-temporal variation of river hydrological parameters is surprisingly poor. In situ gauge stations are limited in spatial and temporal coverage, and their number has been decreasing during the past decades. On the other hand, remote sensing techniques have proven their ability to measure different parameters within the Earth system. Satellite imagery, for instance, can provide variations in river area with appropriate temporal sampling. In this study, we develop an automatic algorithm for water body area monitoring based on maximum a posteriori estimation of Markov random fields. The algorithm considers pixel intensity, spatial correlation between neighboring pixels, and temporal behavior of the water body to extract accurate water masks. We solve this optimization problem using the graph cuts technique. We also measure the uncertainty associated with the determined water masks. Our method is applied over three different river reaches of Niger and Congo rivers with different hydrological characteristics. We validate the obtained river area time series by comparing with in situ river discharge and satellite altimetric water level time series. Along the Niger River, we obtain correlation coefficients of 0.85–0.96 for river reaches and 0.65 for the Congo River, which is demonstrably an improvement over other river mask retrieval algorithms.

Today's enormous amounts of freely available high-resolution satellite imagery provide the demand for effective preprocessing methods. One such preprocessing method needed in many applications utilizing optical satellite imagery from the Landsat and Sentinel-2 archives is mosaicking. Merging hundreds of single scenes into a single satellite data mosaic before conducting analysis such as land cover classification, change detection, or modelling is often a prerequisite. Maintaining the original data structure and preserving metadata for further modelling or classification would be advantageous for many applications. Furthermore, in other applications, e.g., connected to land cover classification creating the mosaic for a specific period matching the phenological state of the phenomena in nature would be beneficial. In addition, supporting in-house and computing centers not directly connected to a specific cloud provider could be a requirement for some institutions or companies. In the current work, we present a method called Geomosaic that meets these criteria and produces analysis-ready satellite data mosaics from Landsat and Sentinel-2 imagery.•The method described produces analysis-ready satellite data mosaics.•The satellite data mosaics contain pixel metadata usable for further analysis.•The algorithm is available as an open-source tool coded in Python and can be used on multiple platforms., Competing Interests: The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper., (© 2023 The Author(s).)

This letter presents a cloud removal method for reconstructing the missing information in cloud-contaminated regions of a high-resolution (HR) optical satellite image (HRI) using two types of auxiliary images, i.e., a low-resolution (LR) optical satellite composite image (LRI) and a synthetic aperture radar (SAR) image. The LRI contributes low-frequency information, and the SAR image contributes high-frequency information for restoring the HRI. The approach is implemented using structure correspondences established by sparse representation. Specifically, two dictionary pairs are trained jointly: One pair is generated from the HRI and LRI gradient image patches, and the other is generated from the HRI and SAR gradient image patches. Experimental reconstructions of cloud-contaminated regions in HR Thematic Mapper images are performed using three types of auxiliary images, i.e., MODIS 16-day composite only, SAR only, and both MODIS composite and SAR, respectively. It is shown that the MODIS composite or the SAR data alone are not sufficient to restore the missing HR information, whereas the combination of the two types of data can provide both low- and high-frequency information. The proposed approach can achieve a highly accurate result and has potential in areas where land-cover change may occur. [ABSTRACT FROM PUBLISHER]

In this paper, we present a method for the multitemporal and contextual classification of georeferenced optical remote sensing images acquired at different epochs and having different geometrical resolutions. The method is based on Conditional Random Fields (CRFs) for contextual classification. The CRF model is expanded by temporal interaction terms that link neighboring epochs via transition probabilities between different classes. In order to be able to deal with data of different resolution, the class structure at different epochs may vary with the resolution. The goal of the multitemporal classification is an improved classification performance at all individual epochs, but also the detection of land-cover changes, possibly using lower resolution data. This paper also contains a comparison of the performance of different models for the interaction potentials. Results are given for two different test sites in Germany, where Ikonos, RapidEye, and Landsat images are available. Our results show that the multitemporal classification does indeed increase the overall accuracy of all epochs compared to a monotemporal classification and to a state-of-the-art multitemporal classification method, and that it is feasible to detect changes in lower resolution images. [ABSTRACT FROM PUBLISHER]

In this paper, we show how satellite images taken by space-borne radar sensors can be used to determine mesoscale high-resolution wind fields in synergy with cloud parameters from optical data and, thus, help in the task of maintenance and planning offshore wind farms. The aim of this paper is to use synthetic aperture radar (SAR) and medium resolution imaging spectrometer (MERIS) onboard the environmental satellite (ENVISAT) in synergy to analyze severe weather systems, in particular, to describe the spatial evolution of the atmospheric boundary layer processes involved in cold air outbreaks. We investigated the fine-scale structure of a severe weather case on November 1, 2006 over the North Sea using satellite data. The satellite data are compared with numerical model results of the German Weather Service ldquoLokal Modellrdquo (LM) and the high-resolution limited area model (HIRLAM). LM and HIRLAM show differences in mesoscale turbulent behavior and coastal shadowing. Maximum wind speeds of up to 25 m/s are measured by SAR and are confirmed by the models. Significant differences are observed in the location of the maxima. High-resolution ENVISAT ASAR measurements provide very detailed information on small-scale atmospheric features, which seem to not be captured well by the analyzed numerical models, in particular, in coastal areas. Meteosat second generation (MSG) is used to determine the movement of cloud patterns. Cloud patterns seen in the optical data and radar cross-section modulation give a consistent dynamical picture of the atmospheric processes. The relevance for offshore wind farming is discussed. [ABSTRACT FROM PUBLISHER]

Slow-moving landslides exhibit persistent but non-uniform motion at low rates which makes them exceptional natural laboratories to study the mechanisms that control the dynamics of unstable hillslopes. Here we leverage 4.5+ years of satellite-based radar and optical remote sensing data to quantify the kinematics of a slow-moving landslide in the tropical rural environment of the Kivu Rift, with unprecedented high spatial and temporal resolution. We measure landslide motion using sub-pixel image correlation methods and invert these data into dense time series that capture weekly to multi-year changes in landslide kinematics. We cross-validate and compare our satellite-based results with very-high-resolution Unoccupied Aircraft System topographic datasets, and explore how rainfall, simulated pore-water pressure, and nearby earthquakes control the overall landslide behaviour. The landslide exhibited seasonal and multi-year velocity variations that varied across the landslide kinematic units. While rainfall-induced changes in pore-water pressure exerts a primary control on the landslide motion, these alone cannot explain the observed variability in landslide behaviour. We suggest instead that the observed landslide kinematics result from internal landslide dynamics, such as extension, compression, material redistribution, and interactions within and between kinematic units. Our study provides, a rare, detailed overview of the deformation pattern of a landslide located in a tropical environment. In addition, our work highlights the viability of sub-pixel image correlation with long time series of radar-amplitude data to quantify surface deformation in tropical environments where optical data is limited by persistent cloud cover and emphasize the importance of exploiting synergies between multiple types of data to capture the complex kinematic pattern of landslides. [Display omitted] • We provide a rare analysis of the dynamics of a landslide in tropical environment. • We used dense time series from radar and optical data to quantify its motion. • Rainfall-induced changes in pore-water pressure exerts the main control. • High-spatio-temporal kinematic data stress the role of internal landslide dynamics. • We highlight the viability of sub-pixel image correlation with radar-amplitude data. [ABSTRACT FROM AUTHOR]

Himalayan glaciers have shrunk rapidly in recent decades, but the spatial pattern of ice loss is highly variable and appears to be modulated by factors relating to individual glacier characteristics. This hinders our ability to predict their future evolution, which is vital for water resource management. The aim of this study is to assess recent glacier changes in the little-studied Annapurna Conservation Area (ACA; area: 7629 km2) in Nepal, and to explore local controls influencing their behaviour. We map changes in glacier area, surface elevation, and ice flow velocity on a large sample of glaciers (n = 162) in the ACA between 2000 and 2016. We found that total glacier area decreased by 8.5% between 2000 and 2014/15. Ice surface velocity changes between 2002 and 2016 were variable, with no clear trend of acceleration or deceleration. The mean surface elevation change for a smaller sample of glaciers (n = 72) was −0.33 ± 0.22 m a−1 between 2000 and 2013/16, which equates to a mean mass balance of −0.28 ± 0.24 m w.e. a−1. There was a trend of increasingly less negative mass balance towards the north. Glaciers that lost the most mass in the north of the ACA tended to have lower maximum elevations, bottom-heavy hypsometries, and were more likely to be avalanche-fed. However, these patterns were not apparent in glaciers in central ACA. There was no significant difference in the mean surface elevation change rate on the ablation zones of debris-covered compared with debris-free glaciers. Our work shows that glaciers in the ACA are losing area and mass at variable rates, but that the influence of local controls is complex, which introduces large uncertainties when predicting their future evolution. [ABSTRACT FROM AUTHOR]