JARS thanks the reviewers who served the journal in 2022.

The high temporal resolution data has a great opportunity for specific crop mapping. The recent advancement in remote sensing data had a good impact on crop classification and mapping. In this research work, modified possibilistic c-means (MPCM) fuzzy machine learning approach has been applied for castor crop mapping, using sentinel 2A/2B satellite temporal images and local convolution as adaptive modified possibilistic local information c-means (ADMPLICM) algorithm was applied to remove isolated unwanted pixels. In training sample parameter “individual sample as mean” method was used to handle heterogeneity within the castor fields. The class-based sensor independent normalized difference vegetation index (CBSI-NDVI), normalized difference vegetation index (Red-NIR), and modified NDVI were produced, then separability analysis was conducted for the selecting optimum temporal date for castor crop mapping. The heterogeneity effect within the field was studied using variance parameters, and its values were observed low within the castor crop field. While considering the “individual sample as mean” training parameter approach on CBSI-NDVI, the variance of the target was decreased from 0.019 to 0.016. The identified castor field variance was close to the training field based on their high membership values and these fields had low mean membership difference (MMD) value. MMD values of castor training class with other crop classes were very high (0.425 – Isabgol, 0.502 – wheat, 0.259 – Fenugreek, 0.809 – Mustard, and 0.569 – Cumin), which indicates other crops were distinguished very well.

For specific crop mapping, temporal data is used to handle the spectral overlap between classes. But a number of times, availability of temporal optical remote sensing data is not consistent with the requirements. In this study, a medicinal crop, psyllium husk, was mapped in Jalore district of Rajasthan, India while trying to integrate optical temporal data with synthetic aperture radar (SAR) data. Vegetation indices derived from optical and SAR data, i.e., modified soil adjusted vegetation index 2 (MSAVI2) and radar vegetation index (RVI), respectively, were used to prepare temporal indices databases. Three different temporal layer combinations of MSAVI2 and RVI layers, corresponding to different phenological stages of the target crop, were studied to determine the efficacy of different sensors for crop mapping. The outputs were assessed on the basis of classification efficiency with mean membership difference and variance for determining the effect of heterogeneity in the mapped fields. The results were compared with those obtained using only optical temporal data. From the results obtained, it was observed that the introduction of SAR-based vegetation index as a temporal layer gave good results for fully vegetated and harvest stages of the target crop. The variance within mapped fields was observed as low as 0.0003.

The leaf area index (LAI) is an important parameter for describing the growth status and canopy structure of vegetation. The rapid and accurate acquisition of the vegetation or agroforestry LAI has great scientific significance in agroforestry ecological ecosystems research and a very important practical value for guiding agricultural and forestry production. In this study, the typical tropical crops (rubber forest) in Hainan Island were selected as the research area, the empirical and neural network (NN) LAI estimation models of rubber forest were constructed based on satellite remote sensing vegetation indices and the field LAI measurement data, and the spatiotemporal variation was analyzed. The results showed that, compared with normalized difference vegetation index (NDVI), green NDVI (GNDVI), ratio VI (RVI), normalized near-infrared (NNIR), wide dynamic range VI (WDRVI), and normalized difference water index (NDWI), enhanced vegetation index (EVI), soil adjusted vegetation index (SAVI), DVI, renormalized DVI (RDVI), and modified SAVI (MSAVI) have higher correlations with LAI. Among the LAI estimation models of rubber forest based on empirical and artificial NN (ANN) models, the estimation accuracy of ANN achieves the highest value. The linear fitting determination coefficient R² of the observed and simulated rubber forest LAI was 0.85 (p < 0.001), the root mean square error (RMSE) was 0.15, and the average relative error (RE) was 1.93%. However, there was underestimation in the middle-value area and overestimation in the high- and low-value areas of LAI. Based on remote sensing mapping of the rubber forest LAI, the high LAI values (4.40 to 6.00 m² m ^{− 2}) were mainly distributed in Danzhou and Baisha (west of Hainan Island); the middle LAI values (3.80 to 4.40 m² m ^{− 2}) were mainly located in Chengmai, Tunchang, and Qiongzhong (middle of Hainan Island); and the low LAI values (<3.80 m² m ^{− 2}) were shown primarily on Ding’an, Qionghai, Wanning, Ledong, and Sanya (east and south of Hainan Island). In summary, the remote sensing estimation model for the rubber plantation LAI based on the vegetation index has high accuracy and good values for application.

Our study aimed to determine the most suitable indicator area and a remote sensing-based index (RSI) for warning winter–spring rice yield in the Central Highlands, Vietnam. Rice yield and RSIs, including vegetation condition index (VCI), temperature condition index, vegetation health index, and temperature-vegetation dryness index, in the period 2000 to 2020 were analyzed. Mann–Kendall nonparametric test and Sen’s slope were used to determine the trends of rice yield and RSIs. Meanwhile, multiple correlation coefficients were used to determine the most suitable indicator area and RSI for rice yield warning, and stepwise linear regression was used to evaluate the reliability of their warning ability. Research results showed that VCI was the most suitable RSI for early rice yield warning. The indicator area identified based on VCI occupied 5% of the area of the Central Highlands. This area had more difficult farming conditions than the rice-growing area and crops grown in it were also sensitive to water sources. Furthermore, the results also indicated that the VCI of the indicator area could predict rice yield from 1 to 3 months before harvest. The forecast accuracy depended on the sensitivity of rice yield to VCI and the stability of rice yield in each province.

Our study established a machine learning (ML) model that could predict the apple yield based on various satellite multisensor data, such as climatological, SAR backscatter, terrain distribution, and soil factors, grouped as 26 subcriteria. A total of 986 apple orchards database were collected from 2018 to 2021 in Kashmir Valley, India covering an area of 277953.7 ha farmland. The novelty of our research is the integration of Google Earth Engine cloud and ML models, namely random forest, support vector machine, extreme gradient boosting, K-nearest neighbors, and Cubist along with the geographic information system/remote sensing technology to create an accurate and comprehensive apple yield prediction model in the precision agriculture realm for highlands. The multicollinearity testing indicated that the tolerance and VIF values of all the conditioning factors were <0.1 and <6.85, respectively, indicating no multicollinearity problems among the apple yield suitability factors. Among the tested ML models, the Cubist model performed best, with R² of 0.83, root-mean-squared error of 0.56 t / ha, and mean absolute error of 0.2 t / ha. The results showed a low mean fruit yield during 2018 of 12.36 ton / ha, whereas maximum fruit yield was reflected in 2021 of 14.05 ton / ha. The heat map revealed the highest normalized differential vegetation index along with vertical-vertical/ vertical-horizontal polarization backscatter, detected during the pre-event of severe snowfall compared to on- and postevent of snowfall for the respective years. Untimely snowing and infestation due to fungi and bacterial diseases regularly reduce fruit yield in the study area. Our study successfully used of high-resolution optical-SAR data combined with ML models as a promising tool for monitoring the yield variability over the highland areas.

The main goal of this study is to evaluate different models for further improvement of the accuracy of land use and land cover (LULC) classification on Google Earth Engine using random forest (RF) and support vector machine (SVM) learning algorithms. Ten indices, namely normalized difference vegetation index, normalized difference soil index, index-based built-up index, biophysical composition index, built-up area extraction index (BAEI), urban index, new built-up index, band ratio for built-up area, bare soil index, and normalized built up area index, were used as input parameters for the machine learning algorithms to improve classification accuracy. The combinatorial analysis of the Sentinel-2 bands and the aforementioned indices allowed us to create four combinations based on surface reflectance characteristics. The study includes data from April 2020 to September 2021 and April 2022 to June 2022. The multitemporal Sentinel-2 data with spatial resolutions of 10 m were used to determine the LULC classification. The major land use classes such as water, forest, grassland, urban areas, and other lands were obtained. Generally, the RF algorithm showed higher accuracy than the SVM. The overall accuracy for RF and SVM was 86.56% and 84.48%, respectively, and the mean Kappa was 0.82 and 0.79, respectively. Using the combination 2 with the RF algorithm and combination 4 with the SVM algorithm for LULC classification was more accurate. The additional use of vegetation indices allowed to increase in the accuracy of LULC classification and separate classes with similar reflection spectra.

Time series interferometry synthetic aperture radar (InSAR) techniques have rarely been applied to detect displacement due to low-magnitude (5.5 ≥ M) earthquakes. This study exploits the combined permanent scatterer (PS) interferometry (PS-InSAR) and differential interferometry (D-InSAR) methods to investigate the deformation rates due to low-magnitude partial series of collapse earthquakes. Kandy, the hill capital of Sri Lanka, was experiencing a series of collapsed earthquakes. Historical geological evidence of dynamic topography with prolonged vertical movement further promotes the initiatory InSAR investigation. A series of Sentinel 1A images from the years 2019 to 2021 were employed. Initial D-InSAR near-zero baseline analysis suggested possible displacement in the range of −10 mm / year and +2 mm / year for subsidence and uplift, respectively. Using these prior motion velocities, the temporal coherence was optimized in PS-InSAR. Variogram models and ordinary Kriging (OK) were employed to predict deformation for the areas with limited PS detection. Descending orbits images PS show a dominant uplift of +20 mm / year, which are concentrated over Kandy urban areas. Movements along the ascending line of sight at Victoria Dam in the south are in the range of −40 mm / year. Slopes along the Mahaweli river lineament to the east show subsidence in the range of −29 to 36 mm / year. The coregistered landslide hazard map of Kandy shows deformation areas are exposed to landslide risk. Changes in terrestrial and ground water levels measured with the gravity recovery and climate experiment/GRACE-Follow-On during the period reveal significant irregularities. The study can be considered a prototype example that can be extended to investigate low-magnitude incomplete collapse earthquakes in different geological and geotechnical setups for ground deformation.

We present a signal-to-noise ratio (SNR) simulator design for high-resolution Earth observation imagers with time-delay and integration (TDI) imaging sensors. SNR is one of the space-borne electro-optic imagers’ driving design and performance evaluation parameters. TDI line imaging sensors are used to meet these imagers’ SNR requirements by increasing effective exposure time during imaging. Determining the optimum number of TDI stages to meet the SNR requirements of the mission is critical for the imaging sensor selection, optical design of the telescope, laboratory radiometric tests, and imaging in orbit. The SNR simulator is based on an end-to-end image chain calculation following the light pathway from the sun to the imager. The simulator comprises radiance, telescope, sensor, and image modules written in the open-access Python environment. The radiance module uses PcModWin/MODerate resolution atmospheric TRANsmission output data of the atmospheric radiance calculation input for the simulator. The simulator computes the SNR for the given telescope and imaging sensor model and simulates associated images to show the radiometric performance. A high-resolution electro-optic imager model is used to demonstrate the performance and versatility of the simulator for the case studies. Experimental simulator validation is demonstrated for the images acquired by the Göktürk-2 space-borne imager.

The constant land use and land cover (LULC) changes combined with climatic factors are frequently assigned to anthropogenic eutrophication, one of the main ecological imbalances in aquatic systems characterized by dense phytoplankton proliferation. Beyond the degradation of freshwater ecosystems, some cyanobacterial and algae species produce toxins harmful to living beings. Distinct studies in the literature, usually supported by in situ data, have discussed the influence of LULC and climatic changes on phytoplankton bloom events. In this context, motivated by the importance of understanding the environmental mechanisms assigned to phytoplankton bloom events and considering the difficulties imposed by field data collection, our study focuses on analyzing the mentioned issue only using remotely sensed time series data. For this purpose, we performed a temporal analysis between 1985 and 2018 over a portion of the Barra Bonita Hydroelectric Reservoir, Brazil. Initially, we obtained the landscape occupation, precipitation, and temperature information from the MapBiomas, FLDAS, and CHIRPS projects, respectively. A fully automatic algorithm fed by Landsat image series and supported by Google Earth Engine functions was developed and employed to identify and quantify phytoplankton bloom events. Then, the obtained data were inspected by distinct statistical procedures, including correlation and trend analysis. Although there was an absence of a relationship between the climatic components and the emergence of phytoplankton blooms, it was identified using linear regression models (R² ≥ 78 % ) an intensification of blooms after the increase in nonnatural forestry areas, reduction of pastures, and advance of agricultural areas. Furthermore, machine learning methods were employed to obtain nonlinear regression models (R² ≥ 73 % ), making evident that the landscape changes are mainly responsible for the phytoplankton insurgences in the analyzed region. This result agrees with other studies found in the literature and highlights the possibility of investigating anthropogenic eutrophication only using remotely sensed data and automatic algorithms.

Remote sensing monitoring of forest carbon storage against the background of climate change is a popular research topic, and multisource remote sensing is significant for improving its accuracy at a regional scale. Improving accuracy at a low cost is a current challenge. In this study, the spaceborne light detection and ranging (lidar) ICESat-2/advanced terrain laser altimeter system (ATLAS), along with an optimized random forest regression (RF) model and 54 sample plots, was used to obtain early estimates of carbon storage at the canopy scale in the footprints of ICESat-2/ATLAS. On this basis, combined with Landsat 8 operational land imager (OLI) data and a deep neural network (DNN) model, a regional-scale remote sensing estimation of forest carbon storage in Shangri-La, a mountainous area in Southwest China, was performed. The coefficient of determination (R²) and root mean square error (RMSE) were used to assess the estimation results. The results showed that (1) the apparent surface reflectance (ASR) and other parameters of the ATLAS spots had a strong correlation with the carbon storage of mountain forests, and the optimized RF model estimated the carbon storage well at the canopy scale. The modeling accuracy was R² = 0.8890, the verification accuracy was R² = 0.7750, and the contribution rate of the ASR was 22.76% in the model. (2) Using the carbon storage of 74,873 ATLAS footprints obtained by the RF model as the training sample for deep learning modeling, along with the Landsat 8 OLI vegetation index, a regional-scale forest carbon storage (DNN) estimation model was constructed. The model verification accuracy was R² = 0.5433 and RMSE = 6.6402 Mg / ha. (3) The estimated population carbon storage based on the DNN model was 4.717 × 10⁷ Mg, with an average value of 49.04 Mg / ha in the study area. The model estimates were valid. In conclusion, the results estimated by a small sample were used as a large sample training dataset for region-scale deep learning, which can be effective in reducing costs.

The suitability of agricultural lands is basic data for both sustainable agricultural planning and optimal orientation of future urbanization projects. Indeed, unplanned urban expansion leads to many social and ecological problems. The prediction of urban expansion will strongly help the decision makers and local authorities to create a suitable future urban planning while preserving the regional ecosystem and especially the agricultural lands. Using multitemporal satellite images acquired in 1997, 2001, and 2013 over the Mitidja plain in Algeria, the aim of this study is the analysis of the land use and land cover change (LULCC) spatio-temporal distributions from 1997 to 2013 to model and then predict the future LULCC in 2025. The cellular automata Markov chain model (CA-Markov) that we developed is used for LULCC prediction with a focus on urban sprawl and agricultural lands degradation over the Mitidja plain. Once calibrated and validated, the CA-Markov model generates land suitability maps that indicate the best locations for urban expansion while safekeeping the agricultural lands. The results are interesting and present valuable tools for urban planners and policymakers to ensure the sustainability management of agricultural areas over the study region.

Reservoirs are aquatic environments with complex interactions between bio-chemical and physical components. Many of those parameters can be estimated from remote sensing tools, especially since increasing technological progress allows the analysis of those environments in high spatial and temporal resolutions. This work aimed to analyze the water quality of four reservoirs, with high energy productivity, located in the center-southern region of Brazil from bio-optical and thermal water parameters. Therefore, hundreds of Landsat 5 (TM), 7 (ETM+), and 8 (OLI) satellite images were processed by the Google Earth Engine cloud computing platform to estimate water surface temperature, chlorophyll-a density (chl-a), Secchi depth (SD), and trophic state index (TSI). To calibrate the equations, 11 in situ monitoring points were used in 27 scenes from the TM and ETM+ sensors and 30 from the OLI sensor, within 6 days interval between the monitoring date and the image acquisition. Additionally, records from two weather stations near the reservoirs were analyzed to evaluate meteorological influences on water parameters. SD has influenced TSI in three reservoirs, while chl-a has acted as a proxy in only one. Moderate correlations were observed between air temperature, rainfall, and global radiation and the bio-optical and thermal water parameters. Remote sensing has detected a clear spatio-temporal variation, which allowed regionalizing the reservoirs. Reservoir regionalization based on the mapped parameters may be able not only to visualize the spatial dynamics of the waters but also as a tool for the management of water bodies with multiple users and issues.

Multifrequency synthetic aperture radar (SAR) data have been applied to discriminate subtle differences in the vegetation and to better characterize its structural properties, since each SAR frequency will interact with the different sections of the vegetation canopy. In this study, our main objective was to evaluate the use of multifrequency Sentinel-1 and ALOS-2/PALSAR-2 data for stem volume estimations in Eucalyptus sp. and Pinus sp. plantations using three different machine learning algorithms: random forest (RF), support vector regression (SVR), and extreme gradient boosting (XGB). Different experiments were carried out using combinations of predictor variables derived from both SAR sensors: backscattering, polarimetric decompositions, and interferometry data, and field data considering specific models for Eucalyptus sp. and Pinus sp. and a generic model comprising all forest plantations data. The machine learning models using predictor variables derived from SAR data achieved moderately high accuracy to predict stem volume, mainly when SAR data were used in combination with stand age (Experiment iv). In the best prediction scenario (Experiment iv), the RF, SVR, and XGB models were able to explain 81.7%, 68.5%, and 81.8% [coefficient of variation (R²) values] of stem volume variability considering the generic models, respectively. Our results pointed out that the RF algorithm showed the best performance in predicting stem volume with significant good results and easier implementation in comparison with the other two algorithms (SVR and XGB).

Remote sensing has become the most effective tool for monitoring and evaluating forest resources. Owing to the development of big data, current research on forest resource information mainly focuses on a regional, national, or global scale. However, there are still certain limitations on small regional scales. We used supervised classification based on the Google Earth Engine cloud platform to extract county land use information and a Whittaker smoother to extract vegetation indices to monitor forest area and vegetation indices in the study areas from the 1980s to the 2020s. The results show that the classification accuracy is improved for the study area where the image quality is affected by clouds after adding complementary datasets of greenness and water bodies. The overall accuracy (OA) increases from 0.82 to 0.89, and the kappa accuracy increases from 0.77 to 0.85. The OA of the classification results was 0.88, and the forest area increased and decreased with an overall increase of 225.35 km². The building area continued to increase with an increase of 34.88 km². The cropland area first decreased and then increased with a complete decrease of 243.6 km². The bare soil area decreased by 16.2 km². The area of the water bodies was relatively stable. The vegetation index indicated good forest growth. Jiangle County’s forest area and forest growth stability cannot be achieved without the protection of national and local policies.

By leaving grain stubble on the field over the winter, farmers can actively contribute to nature conservation. Animals receive food and living space in these areas, endangered herbs too, but at the same time, it is reducing the input of substances into water and function as covered ground to protect the fertile soil from erosion. Remote sensing can deliver information on this type of land cover, which is also important for the agricultural administration. The farmer can be compensated by the government to let stubbles on the fields. The results of this study show that there is a high potential to detect overwintering stubbles fields area for common agriculture policy (CAP)-control to verify agricultural funding with Support Vector Machine (96.5%) by combining remote sensing classification methods with geodata analysis techniques using geoinformation systems (GIS), negative buffering (-20 m), and using threshold settings for correct classified pixel per field. These results give occasion for optimism for the further processing of this land cover by using it as an application for the government to support CAP- and water framework directive-activities.

With the enactment of supportive government policies and the increasing maturity of solar photovoltaic (PV) technologies, solar PV energy has become the most cost-effective new energy resource worldwide. Geospatial information on existing solar PV power systems is necessary to manage and optimize the deployment of new PV facilities. In this study, we propose a new deep-learning network, named the enhanced U-Net (E-UNET), to detect PV facilities from Sentinel-2 multi-spectral remote sensing data. Our E-UNET features an enhanced encoder–decoder structure that can efficiently extract spectral and spatial features simultaneously by combining a multi-spectral three-dimensional convolution path and a multi-scale pooling block. We compare the performance of the E-UNET with other semantic segmentation deep-learning networks and a pixel-based random forest classifier. The experimental results show that the E-UNET performs better than the other methods. It achieves an overall accuracy, Matthews correlation coefficient, F1, kappa coefficient, and recall of 0.989, 0.862, 0.869, 0.934, and 0.875, respectively. The experimental results also indicate that the E-UNET accurately detects PV facilities from various complex environments with high accuracy in terms of PV integrity and details.

Mapping abandoned farmlands' location and spatial pattern is essential for rural planning. Monitoring small abandoned farmland based on dynamic farmland's phenology is challenging to due to conflict between spatial and temporal resolutions of conventional satellite missions. A unique approach that combines satellite constellation imagery with improved automatic radiometric normalization method for more temporally consistent reflectance were proposed. Applying it in Ami-town, Ibaraki Prefecture, Japan, abandoned farmlands were identified with 3-m resolution and a Kappa coefficient of 0.81 based on satellite constellation-based time-series data; however, the discrimination of the degree of abandonment was challenging. Unlike conventional abandoned farmland mapping approaches which targets spatially large area, our approach provides information at small, fragmented and abandoned farmland in East Asia that enables labor-saving monitoring of land use shifts.

Forest canopy density (FCD) is an essential factor in forest resources investigation and plays a vital role in forest resources management. To replace the traditional time-consuming and labor-intensive survey methods, it is of great importance to study how to select a suitable model at specific temporal and spatial scales for rapid and accurate estimation of FCD in large areas. To reduce workload and errors, we present an approach combing Landsat 5 TM data and biophysical analysis algorithms without considering thermal index (TI) information to model FCD in a temperate forest in Northern China. The scaled vegetation index (SVD) and scaled shadow index were used as variables for modeling and mapping FCD. The dimidiate pixel model based on the normalized difference vegetation index was also used to estimate FCD. We compared the forest management inventory data of 2012 with the FCD results from the biophysical analysis model and the dimidiate pixel model. The results showed that the biophysical analysis model (prediction precision P = 93.62 % , coefficient of determination R² = 0.886, root mean square error RMSE = 0.057) has higher accuracy than the dimidiate pixel model (P = 89.92 % , R² = 0.725, RMSE = 0.078) in estimating FCD. This study shows that the biophysical analysis model without TI information has better performance than the dimidiate pixel model, and can be used for monitoring large scale FCD.

Precise coordinate estimation is a fundamental engineering challenge in many mapping applications. Conventional surveying techniques based on global navigation satellite system (GNSS), light detection and ranging (LiDAR), or total stations involve costly equipment and time-consuming methodologies and may suffer from restrictions, such as occlusions and low satellite availability. In this study, a surveying approach, based on a custom-equipped unmanned aerial vehicle (UAV) and ArUco markers distributed in an unknown area as the potential target mapping points, is proposed and evaluated through simulation. The UAV incorporates a real-time kinematic GNSS receiver and a gimbal unit, with a simple camera and an electronic rangefinder module. The system demonstrates a real-time hierarchy targeting system that allows the UAV to engage with the ground targets through the camera, measure corresponding distances, record UAV coordinates, and then perform a multilateration-based target coordinate estimation. To evaluate the flexibility, efficiency, and onboard performance of the proposed target positioning approach, the method was developed as a robotic operating system software package and tested in the Gazebo robotic simulator on an NVIDIA Jetson TX2. Several mapping environments, along with varying flight type scenarios, were created to evaluate the resulting coordinate estimation errors. The achieved positioning accuracy is very promising and demonstrates a possible use for circular flying trajectories. Along these lines, the proposed methodology may pave the way toward a precise surveying alternative, as it may establish the main surveying method for common mapping applications in the near future and even provide coordinate estimations in demanding and, until now, unattainable areas.

This research presents an evaluation of the accuracy and uncertainty of estimates of river discharge made using satellite observed data sources as input to a modified form of Manning’s equation. Conventional U.S. Geological Survey (USGS) streamflow gaging station data and in-situ measurements of width, depth, height, slope, discharge, and velocity from 30 USGS gage sites were used as ground-truth to assess accuracy. This study explores accuracy in relation to the amount of ground truth information available, the number of calibration points available, and the accuracy of the input data. This research indicates that remotely sensed discharge estimates associated with the modified Manning equation may be expected to have an uncertainty in range of 10% overall given a sufficient number of calibration points. The uncertainty associated with the modified Manning algorithm increased markedly for depths <3 meters (m) and for discharges <1000 cubic meters per second (m³ / s) for many rivers after calibration. Rivers that exhibit (1) a wide range of flow conditions, (2) a significant number of dams in the watershed and along the channel, and (3) a high baseflow index are more likely to have relatively large errors overall and particularly at the low end of the streamflow range. Uncertainty in remotely sensed measurements of water-surface elevation (WSE) and width in the expected range (WSE, + / − 10 cm; Width, + / − 15 m) introduces uncertainty in the discharge estimates on the order of 10% and is greatest at the low end of discharge as rivers get shallower and narrower. As WSE and width measurement uncertainty increases, discharge uncertainty increases accordingly. In general, the observation errors are greater than the errors associated with the algorithm for a well-calibrated model (e.g., 20 calibration points).

Impulse radio ultrawide band (IR-UWB) radar possesses several advantages, such as heat flow insensitivity, high resolution, great penetration, good mobility, and lightweight, making it a promising technology for detecting and rescuing individuals trapped in fire scenarios. However, reliable detection and extraction of vital signs of trapped people in the presence of complex fire environment, such as water droplets, smoke, wall occlusions, and reflections, require an optimal background clutter suppression scheme designed specifically for fire scenarios. This study comprehensively evaluates 20 background noise suppression schemes using an exhaustive approach that involves constructing radar echo simulation signals and designing an evaluation method for algorithms. The practical applicability of the improved design solutions was evaluated by conducting small-scale fire experiments in an acrylic chamber. The results indicate that the designed scheme is effective in overcoming the background interference from the combustion environment and is capable of supporting subsequent trajectory identification and vital sign extraction in the fire scene environment.

Forest biomass is an important biophysical parameter, which delivers vital and valuable information about forest health, growth, productivity, carbon cycle monitoring, forest degradation, and its ecosystem. It is an important inconstant for ecological modeling, carbon stock assessment, and climate change. Forest biomass estimation has been progressively investigated, in which the accuracy of results is good enough to estimate accuracy of biomass; therefore, more accurate estimation of biomass is important for refining the precision and its applicability of these techniques. Hyperspectral remote sensing provides more accurate information about vegetation, so with the combination of advanced hyperspectral datasets it may be a better technique to enhance the results and accuracy of spatial biomass. Airborne hyperspectral data of airborne visible infrared imaging spectrometer-next generation data were demonstrated to estimate above ground biomass (AGB) of a tropical dry deciduous forest. Atmospherically resistant vegetation index, simple ratio index (SRI), normalized difference vegetation index, and enhanced vegetation index (EVI) were used to estimate AGB, in which EVI performs better than other vegetation indices with 0.55 R square value. Plant senescence reflectance index was used to estimate the dry and senescence condition of the forest and its correlations were performed with ground biomass and other vegetation indices.

Plant water stress can be detected via remote sensing. The objective of the study was to determine which leaf water index is best for assessing leaf water content from the laboratory standpoint. This study investigated the relationship between equivalent water thicknesses (EWT), gravimetric water content (GWC), and plant water concentration in the 350- to 2500-nm reflectance spectral range. A total of 277 leaf samples taken from ten different plants were used as calibration dataset, and 605 leaves from different plants, including LOPEX93 and ANGERS database, were used for validation. Three specific indices were analyzed: simple ratio, normalized ratio, and double difference (Datt type of index). A regression approach based on the iteration method at 5-nm interval was used for model calibration. Three bands index was found the most suitable and was validated by 605 leaf samples: for the linear regression model, the index is ( R₁₉₁₀ − R₁₃₄₀ ) / ( R₁₉₁₀ − R₁₁₂₅ ) with R² = 0.96 and root mean square error ( RMSE ) = 0.001 ( g / cm² ) and, for nonlinear regression model the index is ( R₁₉₃₀ − R₁₄₂₅ ) / ( R₁₉₃₀ − R₁₃₆₀ ) with R² = 0.95 and RMSE = 0.001 ( g/cm² ) for EWT. The newly proposed indices take advantage of being able to eliminate additional noise created by the leaf surface, making them helpful for agricultural-related research.

An inverse synthetic aperture radar processing-based algorithm is improved for ship imaging in geosynchronous synthetic aperture radars (GEO SARs) to solve the technical challenges including long accumulation time and complex target motion characteristics. The improvement on the existing algorithm mainly involves four aspects. First, the formula for calculating the appropriate time length of subaperture is derived, according to the principle that the accumulated power of the ship’s echo corresponding to the subaperture duration ensures that the signal-to-noise ratio satisfies the requirement of GEO SARs. The derived formula solves the problem that how long the subaperture length should be taken with regards to a long accumulation time of GEO SAR when imaging ships. Second, we propose an optimization criterion for determining the appropriate subaperture time window by ensuring that the maximum number of ship pixels is detected in the subaperture image. The proposed criterion is used to optimize the location of the subaperture time window in the long accumulation time. Third, the SPECAN algorithm used in the scene imaging process is replaced by an improved nonlinear chirp scaling (NLCS) algorithm to improve the imaging accuracy of the scene. Our improvement on the traditional NLCS algorithm focuses on improving precisions of some of the formulas in the traditional algorithm. Furthermore, a translational compensation algorithm is proposed by considering the influence of spatial variance of slant range history of GEO SARs. The proposed translational compensation algorithm improves the compensation accuracy since the spatial variance of different scatterers in the scene is considered. A series of simulation experiments related with ship imaging in GEO SARs were carried out to verify the correctness of the proposed algorithm and the improvement on the imaging quality compared with that of the existing algorithm.

Owing to the inherent characteristics of synthetic aperture radar (SAR) imaging system based on the coherent imaging modality, SAR images are inevitably corrupted by speckle noise, thereby affecting SAR target recognition accuracy. We explore the impact of speckle noise on SAR target recognition performance. Subsequently, we propose a deep attention convolutional network assisted by a multiscale residual despeckling network to improve SAR target recognition accuracy under speckle corruption. Noise information is first learned via a despeckling subnetwork consisting of dual-branch multiscale feature extraction, feature fusion, and adaptive feature channel selection. Then a classification subnetwork with a cross-dimension interaction attention mechanism is designed to realize feature extraction and identity reasoning of SAR targets. Experimental results on the benchmark moving and stationary target acquisition and recognition dataset demonstrate the effectiveness and superiority of the proposed method.

Accurate detection and segmentation of individual trees from unmanned aerial vehicle images is critical for forestry resource surveys and accurate forest management. Deep learning methods have been used for studies of individual tree crown segmentation, classification, and number of trees in mixed coniferous and broad-leaved forests, but the accuracy needs to be improved. Therefore, this study uses BlendMask, a simpler and more efficient algorithm that combines Mask R-CNN and Yolact algorithms to effectively combine instance-level information with semantic information at a finer granularity level, greatly improving crown segmentation accuracy and classification results. Three coniferous species and five broad-leaved species unmanned aerial vehicle images collected from the Jing Yue multispecies ecological forestry site in Changping District, Beijing, were used as the dataset, and the results were compared with Yolact and Mask R-CNN. The results show that the method described in this work has the highest Kappa coefficient (0.89) and overall accuracy (92.14%) in the test set. For segmentation accuracy, coniferous species’ producer’s accuracy was 0.91 to 0.95, whereas that of broad-leaved species was 0.89 to 0.92. For species classification, the F1-score and mean average precision for coniferous species were greater than 91%, whereas those for broad-leaved species were 77.64% to 85.63%. The accuracy of extracting stand density in low and medium canopy density stands was 0.9909 and 0.9422, respectively, whereas that in high canopy density stands was 0.8913. This study shows that the BlendMask model has a good effect in studying the classification of multiple tree species, the segmentation of individual tree crowns, and the statistics of the number of trees in complex forest areas. Compared with broad-leaved forests and high canopy density stands, this model is more suitable for coniferous forest and medium and low canopy density stand scenarios. This study provides an important tool for obtaining more accurate species classification, canopy segmentation, and resource inventory results in complex forest areas.

In recent days, an unrivaled proliferation of autonomous driving becomes an active area of research in the field of smart city applications. Millions of autonomous cars emerge as ubiquitous in smart cities in the foreseeable future. Vehicle detection and tracking is one of the important cases in smart city applications. It clarifies that the efficient scene parsing methods will be a huge potential. There were several traditional methods employed for the detection of object for the purpose of vehicle detection and road segmentation. However, there were some drawbacks such as reduced rate of detection accuracy, improper road segmentations, and missing detections of occluded objects exist. To overcome the existing issues, we propose an effective multi focus object detector (MFOD) that retrieves spatiotemporal data and considers multiple cues of an object for detection and tracking. The proposed MFOD performs lane detection, vehicle detection, and tracking. To accomplish this task, modified Hough transform is proposed for lane detection and road segmentation. Residual attention unit (RAU) with shared convolutional neural network layers is deployed for vehicle detection and tracking. It makes use of motion model to update the network and creates visualization map to explore the visibility of occluded objects and spatial attention map for accurate prediction of object location. The proposed MFOD is also extended for scene parsing task. We performed extensive analysis of the proposed MFOD with standard evaluation metrics on two different datasets, such as KITTI for object detection and tracking task and cityscapes for scene parsing task. The experimental results demonstrate that the proposed MFOD achieves 3% to 4% improvement compared with the existing methods.

Hyperspectral image (HSI) clustering is a challenging task due to its containing rich spectral information and spatial features. Subspace clustering has been used to explore the intrinsic connections between data points successfully. Graph convolutional subspace clustering (GCSC) was used to achieve robust HSI clustering. The model remaps the self-expressions of the data to non-Euclidean domains to generate a robust graph embedding dictionary. The efficient kernel graph convolutional subspace clustering model utilizes a subspace clustering model with the Frobenius norm and a Gaussian kernel function to achieve a globally optimal closed-form solution that is easy to implement, train, and apply. To reduce the impact of image noise on the segmentation effect, we consider spatial nonlocal self-similarity using nonlocal information blocks based on the fact that the spectral features of HSIs lie in low-dimensional subspaces. At the same time, the three-dimensional tensor generated by the nonlocal similar image blocks is estimated using a robust statistic function. Principal component analysis is used to remove the redundant information of hyperspectral data and extract the main features of pixels. Then a robust statistical function and nonlocal information block are used to construct the affinity matrix. The redundant information prevalent in the whole natural image is used for denoising, that is, the image block is taken as the unit to find similar areas in the image, then these regions are weighted and averaged. Since all pixels in the image are used, the noise can be better filtered. Finally, the constructed affinity matrix is applied to spectral clustering to obtain better clustering results. The model used is called nonlocal mean robust statistical kernel graph convolutional subspace clustering (N-RKGCSC). Experimental results on Salinas, Indian Pines, Pavia Center, and Pavia University datasets showed that the N-RKGCSC model could eliminate the interference of additive and multiplicative noise well, achieving better segmentation performance and better noise immunity.

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.

High-resolution synthetic aperture radar (SAR) sensors are now commonly used for flood detection. Automated detection tends to be limited to rural areas owing to the complicated backscattering mechanisms occurring in urban areas. Flooding can be identified in urban areas by searching for increased SAR backscatter in a postflood image due to double scattering between water and adjacent buildings, compared with a preflood image where double scattering is between unflooded ground and buildings. For co-polarized data, if φ is the angle between the building wall and the satellite direction of travel, double scattering is strongest for φ = 0 deg and falls off as φ increases. Theoretical studies estimating the ratio of flooded-to-unflooded double scatterer (DS) radar cross section (RCS) using X-band SAR data, found that the ratio was high at φ = 0 deg but only small at φ > 10 deg. Ostensibly, this implies that few DSs might be detected in an urban area. However, experiments on real images have called into question the veracity of the modeling. We describe an empirical study to examine the relationship between the flooded-to-unflooded DS RCS ratio and φ in Sentinel-1 (S-1) C-band data. We use high-resolution light detection and ranging and aerial photographs so that φ can be measured accurately and is based on S-1 images from flood events that occurred in the United Kingdom during the storms of winter 2019 to 2020. Results indicate that vertical–vertical polarization is better than vertical–horizontal at distinguishing flooded from unflooded DS; that the theoretical model used underestimates the number of DS with high RCS ratios in the φ range 10 deg to 30 deg; and that sufficient DS ground heights can be determined to estimate an accurate local average flood level, although in high density housing there are less of these due to the presence of adjacent buildings.

For hypersonic target detection, the range migration (RM) and Doppler frequency migration (DFM) often occur during coherent processing time, which would lead to integration loss and degrade detection performance. Aiming at these problems, this paper proposes a coherent integration method based on shear transform (ST) and dechirp process (DP), namely ST-DP. Specifically, the RM is eliminated by shearing the echo data plane, and then the DFM is compensated by dechirp process. Eventually, the Fourier transform (FT) is carried out in the slow time dimension to realize coherent integration. Compared with existed methods, the presented method has a coherent integration and detection performance close to that of GRFT without increasing the computational cost, while there is no blind speed sidelobe (BSSL) effect. Some simulation experiments are given to prove the practicality and efficacy of the presented method.

A variable projection angle is obtained based on the geometric algebra (GA) projection operator using the geometric meaning of complex vectors’ inner product. Considered that the inner product processing is widely used in radar applications, i.e., radar signal detection and matched filtering, a geometric detector, and a geometric matched filter are proposed based on the projection angle in the GA framework. Several simulations, as well as experiments, confirm that the proposed geometric detector overperforms the traditional constant false-alarm rate (CFAR) detector, especially in the uniform background, and in adaptions to the clutter edge and large dynamic change of signals. Moreover, the higher efficiency of the geometric detector is indicated by the analysis comparison of computation complexity, which indicated that the geometric detector possesses higher efficiency than the traditional CFAR detectors. The response characteristics of the geometric matched filter for intermittent sampling and repeater jamming (ISRJ) are also validated through theoretical and simulation analysis. The presented results provide a geometric perspective for solving the design of radar detectors and the suppression of ISRJ.

Spectral unmixing (SU) of hyperspectral images (HSIs) is one of the important areas in remote sensing (RS) that needs to be carefully addressed in different RS applications. Despite the high spectral resolution of the hyperspectral data, the relatively low spatial resolution of the sensors may lead to mixture of different pure materials within the image pixels. In this case, the spectrum of a given pixel recorded by the sensor can be a combination of multiple spectra each belonging to a unique material in that pixel. SU is then used as a technique to extract the spectral characteristics of the different materials within the mixed pixels and to recover the spectrum of each pure spectral signature, called endmember. Block-sparsity exists in hyperspectral images as a result of spectral similarity between neighboring pixels. In block-sparse signals, the nonzero samples occur in clusters and the pattern of the clusters is often supposed to be unavailable as prior information. This paper presents an innovative SU approach for HSIs based on block-sparse structure. Hyperspectral unmixing problem is solved using pattern coupled sparse Bayesian learning strategy. To evaluate the performance of the proposed SU algorithm, it is tested on both synthetic and real hyperspectral data and the quantitative results are compared to those of other state-of-the-art methods in terms of abundance angle distance and mean squared error. The achieved results show the superiority of the proposed algorithm over the other competing methods by a significant margin.

With the development of synthetic aperture radar (SAR) technology, more SAR datasets with high resolution and large scale have been obtained. Research using SAR images to detect and monitor marine targets has become one of the most important marine applications. In recent years, deep learning has been widely applied to target detection. However, it was difficult to use deep learning to train an SAR ship detection model in complex scenes. To resolve this problem, an SAR ship detection method combining YOLOv4 and the receptive field block (CY-RFB) was proposed in this paper. Extensive experimental results on the SAR-Ship-Dataset and SSDD datasets demonstrated that the proposed method had achieved supreme detection performance compared to the state-of-the-art ship detection methods in complex scenes, whether they were in offshore or inshore scenes of SAR images.

Inverse synthetic aperture radar (ISAR) autofocus imaging performance is challenged by the residual phase errors that arise from the traditional motion compensation methods with sparse aperture. Moreover, sparse reconstruction algorithms usually ignore the block sparsity in the ISAR image, which cannot recover structured information of the block structure target completely. An imaging method based on joint minimum Tsallis entropy and pattern-coupled sparse Bayesian learning algorithm is proposed to achieve ISAR autofocus imaging of block structure targets with sparse aperture. A pattern-coupled hierarchical complex Gaussian prior is utilized to characterize the correlation among ISAR image pixels in the complex domain. Such a prior model has the potential to encourage block-sparse patterns and achieve ISAR sparse aperture imaging of block structure targets without knowing the block partition prior information. The residual phase errors are estimated based on the minimum Tsallis entropy criterion during the sparse reconstruction of ISAR image to achieve ISAR autofocus imaging, which has the advantage of improving the computational efficiency compared with minimum Shannon entropy criterion. The superiority of the proposed algorithm is verified by the experimental results based on both simulated and measured data.

Multiview synthetic aperture radar (SAR) images could provide richer information for target recognition as reported in previous works. This study exploits multiview SAR images with application to target recognition via the combination of joint sparse representation (JSR) and random weight matrix. First, the multiview SAR images are clustered into several view sets via a coarse clustering algorithm. In this way, the views in each set tend to share high correlations so they can be properly represented with JSR. For the reconstruction errors from different views, they are fused through a random weight matrix, which includes a rich set of linear weight vectors. Afterwards, the fused errors at different choices of weight vectors are analyzed to form a decision value for target recognition, which could better reflect the differences between different classes through multiview SAR images in comparison with traditional JSR-based methods. Therefore, the overall SAR target recognition performance can be effectively enhanced. Experiments are investigated on the moving and stationary target acquisition and recognition dataset under several operating conditions. The results reveal the high effectiveness of the proposed method under standard operating condition and good robustness under typical extended operating conditions including depression angle variance, configuration variants, noise corruption, partial occlusion, and resolution variance.

Imaging through degraded visual environments is a challenging task in remote sensing missions. Image degradation may come from the loss of contrast due to particle scattering and/or distortion due to turbulence-induced effects. The problem is especially challenging when imaging from moving platforms such as autonomous underwater vehicles. One potential approach to address these issues is to use multiple images and employ a multi-frame image fusion technique to aid in the recovery of corrupted image detail and quality. A machine learning (ML)-based image enhancement and fusion technique is investigated to restore images distorted by underwater turbulence. The main contributions include the incorporation of a ML-based image weight predictor to predict the ideal weight maps to be used in an image fusion process. This network is trained using a generative adversarial network framework and a synthetically generated image dataset, which is created according to an analytical image degradation model. In addition, the image loss function for the weight map predictor is determined by the final fused image, resulting in a balanced fusion technique that can reduce image distortion, recover crisp image detail, and reduce the overall noise figure. Another key contribution of this paper is the adoption of an image loss function that incorporates an innovative combined correntropy and Fourier space loss function to reinforce the network in both the spatial and frequency domain. The performance of the proposed algorithm is evaluated using a synthetic validation dataset of images and several real datasets captured at the Naval Research Lab’s Simulated Turbulence and Turbidity Environment under various controlled turbulence intensities.

Obtaining change information in different periods from a pair of registered satellite remote sensing images is of great significance to urban planning, so change detection (CD) technology has attracted extensive attention in recent years. In recent years, convolutional neural networks have set off a boom in many artificial intelligence research fields because of their excellent feature extraction performance. However, the common convolution operation mainly focuses on the abstraction of the semantic information of the features, which often leads to the details of the features being ignored and thus affects the final accuracy. For example, the contour details of changing objects and the structural information of small objects are often lost. We propose a Siamese network that enhances contour and structural details to achieve higher-accuracy CD tasks for bitemporal remote sensing images. In this network, we propose an efficient contour-enhanced convolutional block that is based on the reparameterization technique. The contour-enhanced convolutional block strengthens the extraction of structural and contour features by integrating different branches. In addition, inspired by NestedUNet and to better preserve the original location information of features, we use a dense connection as the feature extractor to obtain refined features of bitemporal images. After that, we use a difference module to calculate the change characteristics of the dual-time image, and we use atrous spatial pyramid pooling and enhanced spatial attention to further refine the obtained change characteristics. We conduct extensive experiments on three different datasets to verify the effectiveness of our model. Experimental results show that our method outperforms state-of-the-art methods in both overall accuracy and visualization details.

Striping effects are common phenomena in remote sensing images, and they significantly limit subsequent applications. Although many destriping approaches have been developed, there aren’t many that can completely eliminate complex stripes with varying levels of strength. To address this issue, we propose a stripe removal model based on a variable weight coefficient and group sparse regularization. Specifically, rather than a single scalar for the stripes in most approaches, different weights are set for different stripe rows to estimate the stripes with varying intensities. An adaptive method to estimate the weight matrix is proposed. On the other hand, group sparsity regularization is employed to constrain the entire stripe. In addition, region weights are designed for regions with different stripe characteristics. The alternating direction multiplier method is employed to solve the proposed model by alternating minimization. Experimental results based on simulation and real data demonstrate that the proposed model outperforms other advanced methods in terms of stripe noise removal and image detail preservation.

Remote sensing image scene classification has been widely researched with the aim of assigning semantics labels to the land cover. Although convolutional neural networks (CNN), such as VggNet and ResNet, have achieved good performance, the complex background and redundant information of remote sensing images restrict the improvement of final accuracy. We propose an enhanced multihead self-attention block network, which effectively reduces the adverse impact of background and emphasize the main information. In this model, due to the possible redundancy of high-level information of CNN, we only replace the final three bottleneck blocks of ResNet50 with the enhanced multihead self-attention layer to focus on the salient region of each image more effectively. Our enhanced multihead self-attention layer provides the following improvements over the classical module. First, we construct a triple-way convolution to deal with the arbitrary directionality of remote sensing images and get more stable attention information. Then, the improved relative position encodings are used to consider the relative distance between different location features. Finally, we use depthwise convolution and InstanceNorm operation to restore the diversity ability of multiheads. The contrast and ablation experiments carried out on three public datasets show our approach improves upon the baseline significantly and achieves remarkable performance compared with some state-of-the-art methods.

The rich context information and multiscale ground information in remote sensing images are crucial to improving the semantic segmentation accuracy. Therefore, we propose a remote sensing image semantic segmentation method that integrates multilevel spatial channel attention and multi-scale dilated convolution, effectively addressing the issue of poor segmentation performance of small target objects in remote sensing images. This method builds a multilevel characteristic fusion structure, combining deep-level semantic characteristics with the details of the shallow levels to generate multiscale feature diagrams. Then, we introduce the dilated convolution of the series combination in each layer of the atrous spatial pyramid pooling structure to reduce the loss of small target information. Finally, using convolutional conditional random field to describe the context information on the space and edges to improve the model’s ability to extract details. We prove the effectiveness of the model on the three public datasets. From the quantitative point of view, we mainly evaluate the four indicators of the model’s F1 score, overall accuracy (OA), Intersection over Union (IoU), and Mean Intersection over Union (MIoU). On GID dataset, F1 score, OA, and MIoU reach 87.27, 87.80, and 77.70, respectively, superior to most mainstream semantic segmentation networks.

Data from ocean color monitoring sensors at different spectral channels are available for remote sensing of radiation as seen in the given spectral windows, which is used for deriving information on various atmospheric parameters. However, recent studies have demonstrated the potential of hyperspectral (HS) data over multispectral ocean color (MSOC) data in accurately estimating phytoplankton concentration and in monitoring the coastal dynamics. We propose system spectral shape factor (SSSF)-based approach to recover the embedded HS top-of-atmosphere (TOA) radiance (TOARAD) from the MSOC data. SSSF is defined as convolution of normalized input spectrum and sensor spectral response function (SRF). The advantage of SSSF is that it decouples magnitude and spectral shape part of sensor output and enables recovery of TOARAD. To test this method, the airborne visible/infrared imaging spectrometer-next generation data are used to simulate inputs to MSOC. SRF of ocean color monitor simulated MSOC. SSSF of TOARAD is estimated using SSSF of model-based path radiance spectrum of the pixel, which is similar in spectral shape. Methodology, developed using data from five stations, is validated with data from other five stations. The procedure is successfully repeated using SRFs of sea-viewing wide-field-of-view sensor. The recovered HS data are found to be consistent with the original spectra with very small deviations in spectral angle map (<0.012 rad) spectral information divergence (<5.8 × ^{10 − 5}), mean percentage relative error (MPRE) of TOARAD (<0.7 % ), and MPRE of TOA water leaving radiance (<5.8 % ). This approach possibly opens up research for application of HS analysis on MSOC recovered spectra and for optimization of sensor configurations.

There is a need for a reliable diagnostic process that can provide the information required to maintain optimal operating conditions of phased array antennas. The phase and amplitude information can be retrieved from the complex radiated field. However, phase measurement is laborious and costly at high frequencies. To overcome this challenge, it is of great relevance to create diagnostics methods that solely utilize the amplitude data. The shape and characteristics of the radiated field of a phased array antenna are predicated on the number of operating elements in the array. This implies that a failed element in the array will change the nature of the radiated pattern. We can locate a failed element using a learning algorithm to map the location of a failed element to the radiated field. We propose a technique for finding failed elements using only amplitude data by dividing the array into multiple subarrays. Each subarray is trained with dedicated deep convolutional neural networks for faulty element identification. We evaluated the proposed approach using simulated data for 20 × 20 and 30 × 30 array structures. The results demonstrate that the proposed approach can effectively diagnose and locate failed elements in a phased array antenna using amplitude-only data.

Panax notoginseng is second only to ginseng Panax notoginseng in the Chinese traditional medicine industry. It has extremely high economic value and medicinal value. However, due to the unsustainable planting characteristics of P. notoginseng. The original planting area of P. notoginseng has been exhausted, so the planting area of P. notoginseng has gradually expanded to other areas, and the planting area of P. notoginseng is also changing. Based on Sentinel-2 and Landsat 8 OLI remote sensing image data from 2018 to 2021, the study used Google Earth engine and Random Forest classification to analyze the spatial distribution information of P. notoginseng planting areas in Kaiyuan City, Honghe Prefecture. The results show that, based on Sentinel-2 data, Landsat 8 OLI data, and other multisource data, comprehensively considering the image spectral characteristics and habitat characteristics of P. notoginseng, and applying the Random Forest classification method. The spatial distribution information of P. notoginseng planting areas can be quickly extracted. After calculation, the transfer matrix indicates that the accuracy of the results is high and can meet the requirements of research and application. According to the analysis of the results, the cultivated area of P. notoginseng showed no significant change in 2018 and 2019. It shows an increasing trend during this period. In 2020, the cultivated area of P. notoginseng decreased slightly. While in 2021, the cultivated area of P. notoginseng increased sharply, and the cultivated area of P. notoginseng was more than twice that of the previous year. However, the price of P. notoginseng has shown a trend of decline year by year. By monitoring the planting area of P. notoginseng and analyzing the relationship between the planting area and the price of P. notoginseng, we can predict the future market price of P. notoginseng and plan the rational planting of P. notoginseng.

Soil salinization, one of the important factors leading to global land degradation, seriously affects sustainable agricultural development. Accurate and frequent monitoring of soil salt content (SSC) contributes to the management and restoration of salinized soil. Our study aimed to evaluate the potential to synthetically estimate the salinity in bare soil with different remote sensing sensors (medium-resolution Sentinel-1 and Sentinel-2). The salinity data of 134 surface soil samples were collected in the Shahaoqu irrigation area (SIA) in the Hetao Irrigation District of Inner Mongolia. Simultaneously, the remote sensing images of Sentinel-1 and Sentinel-2 of SIA were obtained. A total of 46 predictors, including 12 remote sensing data (10 multispectral bands, 1 VV, and 1 VH), 8 polarization combination indices, 16 salinity spectral indices, and 10 texture features, were obtained and calculated from the remote sensing image data. Three machine learning algorithms, random forest (RF), support vector machine (SVM), and extreme learning machine (ELM), were used for the construction of soil salinity prediction models based on different combinations of the 46 predictors. The results showed that the remote sensing image data from multiple sensors could be used to estimate soil salinity more accurately than that of a single sensor. Among the three regression models (RF, SVM, and ELM) in our study, RF was the best prediction model (Rv2=0.71, RMSE_v = 0.140 % , and MAE = 0.094 % ). The RF model combining Sentinel-1 SAR and Sentinel-2 multispectral imagery was the most effective in revealing the spatial distribution of soil salinity. The importance analysis of the predictor variables showed that texture features were the main explanatory variables for SSC prediction, with contrast being the most important predictor. The results of our study verified the potential to predict SSC by combining multisource remote sensing data and machine learning models in arid and semiarid regions, which could provide some reference for soil salinization monitoring and control in the bare soil.

Mountain glaciers and alpine lakes are significant indicators of regional climate change. The temperature rise has resulted in the retreat of glaciers, further influencing the ecological cycle and the water balance. The present work mainly focuses on recent glacier and lake elevation changes in the western Kunlun Mountains. The CryoSat-2 satellite altimetry with relatively dense ground track spacing can measure more elevation samples for small glaciers and lakes surrounded by complex topography. The CryoSat-2 data with various algorithms were employed to monitor the water levels of the three typical lakes and the surface elevation changes of their supplying glaciers. The glacier surface elevations showed a slightly increasing trend and remarkable spatial heterogeneity from 2010 to 2021. Meanwhile, the Bangda Lake and Aksayqin Lake displayed clear and strong increasing trends over the observation period with an annual change rate of +0.65 and +0.20 m / yr, respectively. Surprisingly, Gozha Lake exhibited a decreasing trend with the rate of −0.04 m / yr. Regional climate warming led to more melt-water from the glaciers and permafrost flowing into the three typical lakes. The increase in precipitation played a leading role in lake expansion over the western Kunlun Mountains. The present work demonstrated that the satellite altimetry data provide important supplementary data to help us perform the cryosphere research in the data-scarce northern Tibetan Plateau depopulated zone.

As humans increasingly settle in dense urban areas, localized natural and anthropogenic shocks become more likely to impact larger numbers of individuals. Research suggests that resilience to shocks is a function of physical fortifications and social processes including critical infrastructure, social networks, and trust. Although physical fortifications are relatively easy to identify and catalog, social processes elude simple measurement due to data limitations and geographic constraints. Recent work has shown that certain types of infrastructure may correlate with social processes that enhance community resilience; however, the ability to assess where and to what extent that infrastructure exists depends on a complete representation of the built environment. OpenStreetMap (OSM) and Google Places are two sources of data commonly used to locate and characterize infrastructure, but they are often incomplete. We address this limitation by applying a convolutional neural network (CNN) to remote sensing data from Sentinel-2 to estimate the density and type of infrastructure. We compare the classification results to known infrastructure locations from OSM data. Our results show that the CNN classifier performs well and may be used to augment incomplete datasets for a deeper understanding of the prevalence of infrastructure associated with social processes that enhance community resilience.

Deep learning semantic segmentation algorithms have provided improved frameworks for the automated production of land use and land cover (LULC) maps, which significantly increases the frequency of map generation as well as consistency of production quality. In this research, a total of 28 different model variations were examined to improve the accuracy of LULC maps. The experiments were carried out using Landsat 5/7 or Landsat 8 satellite images with the North American land change monitoring system (NALCMS) labels. The performance of various convolutional neural networks and extension combinations were assessed, where Visual Geometry Group Network with an output stride of 4, and modified U-Net architecture, provided the best results. Additional expanded analysis of the generated LULC maps was also provided. Using a deep neural network, this work achieved 92.4% accuracy for 13 LULC classes within southern Manitoba representing a 15.8% improvement over published results for the NALCMS. Based on the large regions of interest, higher radiometric resolution of Landsat 8 data resulted in better overall accuracies (88.04%) compare to Landsat 5/7 (80.66%) for 16 LULC classes. This represents an 11.44% and 4.06% increase in overall accuracy compared to previously published NALCMS results, including larger land area and higher number of LULC classes incorporated into the models compared to other published LULC map automation methods.