Next Issue
Volume 12, June-1
Previous Issue
Volume 12, May-1

Remote Sens., Volume 12, Issue 10 (May-2 2020) – 162 articles

Cover Story (view full-size image): Currently, a growing demand exists for individual building mapping in regions of rapid urban growth in less-developed countries. Here, we present a new convolutional neural network architecture (CNN) called U-net-id specifically designed for the instance segmentation of buildings in very high-resolution satellite imagery. The architecture is built to work with image in tiles, such as remote sensing images, and to reduce the minimum information that needs to be provided by the user—only the images for prediction or only the images and the labelled masks for training. The U-net-id architecture achieves great performance in instance segmentation of the instance building dataset of the city of Joanópolis-SP, Brazil, with a median Intersection over Union of 0.694. Due to the model architecture simplicity, it could easily be reproduced and used for buildings or tested for instance segmentation [...] Read more.
Order results
Result details
Select all
Export citation of selected articles as:
Open AccessArticle
Micro-Pulse Lidar Cruising Measurements in Northern South China Sea
Remote Sens. 2020, 12(10), 1695; https://doi.org/10.3390/rs12101695 - 25 May 2020
Cited by 1 | Viewed by 1034
Abstract
A shipborne micro-pulse lidar (Sigma Space Mini-MPL) was used to measure aerosol extinction coefficient over the northern region of the South China Sea from 9 August to 7 September 2016, the first time a mini-MPL was used for aerosol observation over the cruise [...] Read more.
A shipborne micro-pulse lidar (Sigma Space Mini-MPL) was used to measure aerosol extinction coefficient over the northern region of the South China Sea from 9 August to 7 September 2016, the first time a mini-MPL was used for aerosol observation over the cruise region. The goal of the experiment was to investigate if the compact and affordable mini-MPL was usable for aerosol observation over this region. The measurements were used to calculate vertical profiles of volume extinction coefficient, depolarization ratio, and atmospheric boundary layer height. Aerosol optical depth (AOD) was lower over the southwest side of the cruise region, compared to the northeast side. Most attenuation occurred below 3.5 km, and maximum extinction values over coastal areas were generally about double of values offshore. The extinction coefficients at 532 nm (aerosol and molecular combined) over coastal and offshore areas were on average 0.04 km−1 and 0.02 km−1, respectively. Maximum values reached 0.2 km−1 and 0.14 km−1, respectively. Vertical profiles and back-trajectory calculations indicated vertical and horizontal layering of aerosols from different terrestrial sources. The mean volume depolarization ratio of the aerosols along the cruise was 0.04. The mean atmospheric boundary layer height along the cruise was 653 m, with a diurnal cycle reaching its mean maximum of 1041 m at 12:00 local time, and its mean minimum of 450 m at 20:00 local time. Unfortunately, only 11% of the measurements were usable. This was due to ship instability in rough cruise conditions, lack of stabilization rig, water condensation attached to the eye lens, and high humidity attenuating the echo signal. We recommend against the use of the mini-MPL in this cruise region unless substantial improvements are made to the default setup, e.g., instrument stabilization, instrument protection cover, and more theoretical work taking into account atmospheric gas scattering or absorption. Full article
(This article belongs to the Special Issue Earth Observations in Asia-Oceania)
Show Figures

Graphical abstract

Open AccessArticle
Fuzzy Superpixels Based Semi-Supervised Similarity-Constrained CNN for PolSAR Image Classification
Remote Sens. 2020, 12(10), 1694; https://doi.org/10.3390/rs12101694 - 25 May 2020
Cited by 3 | Viewed by 921
Abstract
Recently, deep learning has been highly successful in image classification. Labeling the PolSAR data, however, is time-consuming and laborious and in response semi-supervised deep learning has been increasingly investigated in PolSAR image classification. Semi-supervised deep learning methods for PolSAR image classification can be [...] Read more.
Recently, deep learning has been highly successful in image classification. Labeling the PolSAR data, however, is time-consuming and laborious and in response semi-supervised deep learning has been increasingly investigated in PolSAR image classification. Semi-supervised deep learning methods for PolSAR image classification can be broadly divided into two categories, namely pixels-based methods and superpixels-based methods. Pixels-based semi-supervised methods are liable to be affected by speckle noises and have a relatively high computational complexity. Superpixels-based methods focus on the superpixels and ignore tiny detail-preserving represented by pixels. In this paper, a Fuzzy superpixels based Semi-supervised Similarity-constrained CNN (FS-SCNN) is proposed. To reduce the effect of speckle noises and preserve the details, FS-SCNN uses a fuzzy superpixels algorithm to segment an image into two parts, superpixels and undetermined pixels. Moreover, the fuzzy superpixels algorithm can also reduce the number of mixed superpixels and improve classification performance. To exploit unlabeled data effectively, we also propose a Similarity-constrained Convolutional Neural Network (SCNN) model to assign pseudo labels to unlabeled data. The final training set consists of the initial labeled data and these pseudo labeled data. Three PolSAR images are used to demonstrate the excellent classification performance of the FS-SCNN method with data of limited labels. Full article
Show Figures

Figure 1

Open AccessArticle
Detection of Grapevine Leafroll-Associated Virus 1 and 3 in White and Red Grapevine Cultivars Using Hyperspectral Imaging
Remote Sens. 2020, 12(10), 1693; https://doi.org/10.3390/rs12101693 - 25 May 2020
Cited by 3 | Viewed by 1130
Abstract
Grapevine leafroll disease (GLD) is considered one of the most widespread grapevine virus diseases, causing severe economic losses worldwide. To date, six grapevine leafroll-associated viruses (GLRaVs) are known as causal agents of the disease, of which GLRaV-1 and -3 induce the strongest symptoms. [...] Read more.
Grapevine leafroll disease (GLD) is considered one of the most widespread grapevine virus diseases, causing severe economic losses worldwide. To date, six grapevine leafroll-associated viruses (GLRaVs) are known as causal agents of the disease, of which GLRaV-1 and -3 induce the strongest symptoms. Due to the lack of efficient curative treatments in the vineyard, identification of infected plants and subsequent uprooting is crucial to reduce the spread of this disease. Ground-based hyperspectral imaging (400–2500 nm) was used in this study in order to identify white and red grapevine plants infected with GLRaV-1 or -3. Disease detection models have been successfully developed for greenhouse plants discriminating symptomatic, asymptomatic, and healthy plants. Furthermore, field tests conducted over three consecutive years showed high detection rates for symptomatic white and red cultivars, respectively. The most important detection wavelengths were used to simulate a multispectral system that achieved classification accuracies comparable to the hyperspectral approach. Although differentiation of asymptomatic and healthy field-grown grapevines showed promising results further investigations are needed to improve classification accuracy. Symptoms caused by GLRaV-1 and -3 could be differentiated. Full article
Show Figures

Graphical abstract

Open AccessArticle
Exploring the Spatial Characteristics of Typhoon-Induced Vegetation Damages in the Southeast Coastal Area of China from 2000 to 2018
Remote Sens. 2020, 12(10), 1692; https://doi.org/10.3390/rs12101692 - 25 May 2020
Cited by 2 | Viewed by 837
Abstract
The southeast coastal area of China (SCAC), a typhoon-prone area with a long coastline, suffers severe damage from typhoons almost every year. Exploring the spatial characteristics of historical typhoon-induced vegetation damage (VD) is crucial to predicting VD after severe typhoon landfalls and improving [...] Read more.
The southeast coastal area of China (SCAC), a typhoon-prone area with a long coastline, suffers severe damage from typhoons almost every year. Exploring the spatial characteristics of historical typhoon-induced vegetation damage (VD) is crucial to predicting VD after severe typhoon landfalls and improving strategies for vegetation protection and restoration. Remote sensing is an efficient and feasible approach for measuring large-scale VD caused by natural disasters. This paper, by exploring the spatial distribution of VD of every severe landfalling typhoon with Google Earth Engine (GEE), aims to reveal the spatial characteristics of typhoon-induced VD in SCAC. Firstly, the values of disaster vegetation damage index (DVDI), difference in enhanced vegetation index (DEVI), and normalized difference vegetation index (DNDVI) for the 28 selected landing typhoons in SCAC were calculated and compared by using moderate resolution imaging spectroradiometer (MODIS) data in GEE. Secondly, every DVDI image was overlaid with land cover, elevation, relative aspect and typhoon path layers in ArcGIS. Thirdly, spatial characteristics of VD were revealed with the aid of spatial statistical analysis. The study found that: (1) DVDI is a more effective index for evaluating VD caused by typhoons. (2) The Pearl River Delta is the most severe VD region. The severe VD regions for four typhoon groups have significantly spatial correlation with typhoon-landing locations. (3) Forests are ranked the first in terms of damaged areas by typhoon in every year, followed by sparse forests. (4) Topography has no influence on VD by a single typhoon event, and relative aspect has no correlation with VD caused by typhoons in SCAC. Full article
(This article belongs to the Section Remote Sensing in Agriculture and Vegetation)
Show Figures

Graphical abstract

Open AccessArticle
The Generation of Soil Spectral Dynamic Feedback Using Landsat 8 Data for Digital Soil Mapping
Remote Sens. 2020, 12(10), 1691; https://doi.org/10.3390/rs12101691 - 25 May 2020
Cited by 1 | Viewed by 883
Abstract
The soil spectral dynamic feedback captured from high temporal resolution remote sensing data, such as MODIS, during the soil drying process after a rainfall could assist with digital soil mapping. However, this method is ineffective in utilizing the images with a relatively high [...] Read more.
The soil spectral dynamic feedback captured from high temporal resolution remote sensing data, such as MODIS, during the soil drying process after a rainfall could assist with digital soil mapping. However, this method is ineffective in utilizing the images with a relatively high spatial resolution. There are an insufficient number of images in the soil drying process since those high spatial resolution images tend to have a low temporal resolution. This study is aimed at generating soil spectral dynamic feedback by integrating the feedback captured from the images with a high spatial resolution during the process of multiple drying after different rainfall events. The Landsat 8 data with a temporal resolution of 16 day was exemplified. Each single spectral feedback obtained from Landsat 8 was first adjusted to eliminate the impact of different rainfall magnitudes. Then, the soil spectral dynamic feedback was reorganized and generated based on the adjusted feedback. Finally, the soil spectral dynamic feedback generated based on Landsat 8 was used for mapping topsoil texture and compared with the mapping results based on the MODIS data and the fusion data of MODIS and Landsat 8. As revealed by the results, not only could the generated soil spectral dynamic feedback based on Landsat 8 data improve the details of the spatial distribution of soil texture, but it also enhances the accuracy of mapping. The mapping accuracy based on Landsat 8 data is higher than that based on the MODIS data and fusion data. The improvements of accuracy are more obvious in the areas with more complex surface conditions. This study widens the scope of application for soil spectral dynamic feedback and provides support for large-scale and high-precision digital soil mapping. Full article
Show Figures

Graphical abstract

Open AccessArticle
Mapping the Global Mangrove Forest Aboveground Biomass Using Multisource Remote Sensing Data
Remote Sens. 2020, 12(10), 1690; https://doi.org/10.3390/rs12101690 - 25 May 2020
Cited by 6 | Viewed by 1555
Abstract
Mangrove forest ecosystems are distributed at the land–sea interface in tropical and subtropical regions and play an important role in carbon cycles and biodiversity. Accurately mapping global mangrove aboveground biomass (AGB) will help us understand how mangrove ecosystems are affected by the impacts [...] Read more.
Mangrove forest ecosystems are distributed at the land–sea interface in tropical and subtropical regions and play an important role in carbon cycles and biodiversity. Accurately mapping global mangrove aboveground biomass (AGB) will help us understand how mangrove ecosystems are affected by the impacts of climatic change and human activities. Light detection and ranging (LiDAR) techniques have been proven to accurately capture the three-dimensional structure of mangroves and LiDAR can estimate forest AGB with high accuracy. In this study, we produced a global mangrove forest AGB map for 2004 at a 250-m resolution by combining ground inventory data, spaceborne LiDAR, optical imagery, climate surfaces, and topographic data with random forest, a machine learning method. From the published literature and free-access datasets of mangrove biomass, we selected 342 surface observations to train and validate the mangrove AGB estimation model. Our global mangrove AGB map showed that average global mangrove AGB density was 115.23 Mg/ha, with a standard deviation of 48.89 Mg/ha. Total global AGB storage within mangrove forests was 1.52 Pg. Cross-validation with observed data demonstrated that our mangrove AGB estimates were reliable. The adjusted coefficient of determination (R2) and root-mean-square error (RMSE) were 0.48 and 75.85 Mg/ha, respectively. Our estimated global mangrove AGB storage was similar to that predicted by previous remote sensing methods, and remote sensing approaches can overcome overestimates from climate-based models. This new biomass map provides information that can help us understand the global mangrove distribution, while also serving as a baseline to monitor trends in global mangrove biomass. Full article
(This article belongs to the Special Issue Remote Sensing in Mangroves)
Show Figures

Graphical abstract

Open AccessArticle
Ubiquitous GIS-Based Forest Fire Susceptibility Mapping Using Artificial Intelligence Methods
Remote Sens. 2020, 12(10), 1689; https://doi.org/10.3390/rs12101689 - 25 May 2020
Cited by 6 | Viewed by 965
Abstract
This study aimed to prepare forest fire susceptibility mapping (FFSM) using a ubiquitous GIS and an ensemble of adaptive neuro fuzzy interface system (ANFIS) with genetic (GA) and simulated annealing (SA) algorithms (ANFIS-GA-SA) and an ensemble of radial basis function (RBF) with an [...] Read more.
This study aimed to prepare forest fire susceptibility mapping (FFSM) using a ubiquitous GIS and an ensemble of adaptive neuro fuzzy interface system (ANFIS) with genetic (GA) and simulated annealing (SA) algorithms (ANFIS-GA-SA) and an ensemble of radial basis function (RBF) with an imperialist competitive algorithm (ICA) (RBF-ICA) model in Chaharmahal and Bakhtiari Province, Iran. The forest fire areas were determined using MODIS satellite imagery and a field survey. The modeling and validation of the models were performed with 70% (183 locations) and 30% (79 locations) of forest fire locations (262 locations), respectively. In order to prepare the FFSM, 10 criteria were then used, namely altitude, rainfall, slope angle, temperature, slope aspect, wind effect, distance to roads, land use, distance to settlements and soil type. After the FFSM was prepared, the maps were designed and implemented for web GIS and mobile application. A receiver operating characteristic (ROC)- area under the curve (AUC) index was used to validate the prepared maps. The ROC-AUC results showed an accuracy of 0.903 for the ANFIS-GA-SA model and an accuracy of 0.878 for the RBF-ICA model. The results of the spatial autocorrelation showed that the occurrence of fire in the study area has a cluster distribution and most of the spatial dependence is related to the distance to settlement, soil and rainfall variables. Full article
Show Figures

Graphical abstract

Open AccessReview
Change Detection Based on Artificial Intelligence: State-of-the-Art and Challenges
Remote Sens. 2020, 12(10), 1688; https://doi.org/10.3390/rs12101688 - 25 May 2020
Cited by 11 | Viewed by 2907
Abstract
Change detection based on remote sensing (RS) data is an important method of detecting changes on the Earth’s surface and has a wide range of applications in urban planning, environmental monitoring, agriculture investigation, disaster assessment, and map revision. In recent years, integrated artificial [...] Read more.
Change detection based on remote sensing (RS) data is an important method of detecting changes on the Earth’s surface and has a wide range of applications in urban planning, environmental monitoring, agriculture investigation, disaster assessment, and map revision. In recent years, integrated artificial intelligence (AI) technology has become a research focus in developing new change detection methods. Although some researchers claim that AI-based change detection approaches outperform traditional change detection approaches, it is not immediately obvious how and to what extent AI can improve the performance of change detection. This review focuses on the state-of-the-art methods, applications, and challenges of AI for change detection. Specifically, the implementation process of AI-based change detection is first introduced. Then, the data from different sensors used for change detection, including optical RS data, synthetic aperture radar (SAR) data, street view images, and combined heterogeneous data, are presented, and the available open datasets are also listed. The general frameworks of AI-based change detection methods are reviewed and analyzed systematically, and the unsupervised schemes used in AI-based change detection are further analyzed. Subsequently, the commonly used networks in AI for change detection are described. From a practical point of view, the application domains of AI-based change detection methods are classified based on their applicability. Finally, the major challenges and prospects of AI for change detection are discussed and delineated, including (a) heterogeneous big data processing, (b) unsupervised AI, and (c) the reliability of AI. This review will be beneficial for researchers in understanding this field. Full article
(This article belongs to the Special Issue Land Use/Cover Change Detection with Geospatial Technologies)
Show Figures

Graphical abstract

Open AccessArticle
Random Forest Spatial Interpolation
Remote Sens. 2020, 12(10), 1687; https://doi.org/10.3390/rs12101687 - 25 May 2020
Cited by 9 | Viewed by 2631
Abstract
For many decades, kriging and deterministic interpolation techniques, such as inverse distance weighting and nearest neighbour interpolation, have been the most popular spatial interpolation techniques. Kriging with external drift and regression kriging have become basic techniques that benefit both from spatial autocorrelation and [...] Read more.
For many decades, kriging and deterministic interpolation techniques, such as inverse distance weighting and nearest neighbour interpolation, have been the most popular spatial interpolation techniques. Kriging with external drift and regression kriging have become basic techniques that benefit both from spatial autocorrelation and covariate information. More recently, machine learning techniques, such as random forest and gradient boosting, have become increasingly popular and are now often used for spatial interpolation. Some attempts have been made to explicitly take the spatial component into account in machine learning, but so far, none of these approaches have taken the natural route of incorporating the nearest observations and their distances to the prediction location as covariates. In this research, we explored the value of including observations at the nearest locations and their distances from the prediction location by introducing Random Forest Spatial Interpolation (RFSI). We compared RFSI with deterministic interpolation methods, ordinary kriging, regression kriging, Random Forest and Random Forest for spatial prediction (RFsp) in three case studies. The first case study made use of synthetic data, i.e., simulations from normally distributed stationary random fields with a known semivariogram, for which ordinary kriging is known to be optimal. The second and third case studies evaluated the performance of the various interpolation methods using daily precipitation data for the 2016–2018 period in Catalonia, Spain, and mean daily temperature for the year 2008 in Croatia. Results of the synthetic case study showed that RFSI outperformed most simple deterministic interpolation techniques and had similar performance as inverse distance weighting and RFsp. As expected, kriging was the most accurate technique in the synthetic case study. In the precipitation and temperature case studies, RFSI mostly outperformed regression kriging, inverse distance weighting, random forest, and RFsp. Moreover, RFSI was substantially faster than RFsp, particularly when the training dataset was large and high-resolution prediction maps were made. Full article
(This article belongs to the Special Issue Machine Learning Methods for Environmental Monitoring)
Show Figures

Graphical abstract

Open AccessArticle
Robust Visual-Inertial Integrated Navigation System Aided by Online Sensor Model Adaption for Autonomous Ground Vehicles in Urban Areas
Remote Sens. 2020, 12(10), 1686; https://doi.org/10.3390/rs12101686 - 25 May 2020
Cited by 3 | Viewed by 916
Abstract
The visual-inertial integrated navigation system (VINS) has been extensively studied over the past decades to provide accurate and low-cost positioning solutions for autonomous systems. Satisfactory performance can be obtained in an ideal scenario with sufficient and static environment features. However, there are usually [...] Read more.
The visual-inertial integrated navigation system (VINS) has been extensively studied over the past decades to provide accurate and low-cost positioning solutions for autonomous systems. Satisfactory performance can be obtained in an ideal scenario with sufficient and static environment features. However, there are usually numerous dynamic objects in deep urban areas, and these moving objects can severely distort the feature-tracking process which is critical to the feature-based VINS. One well-known method that mitigates the effects of dynamic objects is to detect vehicles using deep neural networks and remove the features belonging to surrounding vehicles. However, excessive feature exclusion can severely distort the geometry of feature distribution, leading to limited visual measurements. Instead of directly eliminating the features from dynamic objects, this study proposes to adopt the visual measurement model based on the quality of feature tracking to improve the performance of the VINS. First, a self-tuning covariance estimation approach is proposed to model the uncertainty of each feature measurement by integrating two parts: (1) the geometry of feature distribution (GFD); (2) the quality of feature tracking. Second, an adaptive M-estimator is proposed to correct the measurement residual model to further mitigate the effects of outlier measurements, like the dynamic features. Different from the conventional M-estimator, the proposed method effectively alleviates the reliance on the excessive parameterization of the M-estimator. Experiments were conducted in typical urban areas of Hong Kong with numerous dynamic objects. The results show that the proposed method could effectively mitigate the effects of dynamic objects and improved accuracy of the VINS is obtained when compared with the conventional VINS method. Full article
Show Figures

Graphical abstract

Open AccessArticle
Classification of Arrhythmia by Using Deep Learning with 2-D ECG Spectral Image Representation
Remote Sens. 2020, 12(10), 1685; https://doi.org/10.3390/rs12101685 - 25 May 2020
Cited by 9 | Viewed by 1280
Abstract
The electrocardiogram (ECG) is one of the most extensively employed signals used in the diagnosis and prediction of cardiovascular diseases (CVDs). The ECG signals can capture the heart’s rhythmic irregularities, commonly known as arrhythmias. A careful study of ECG signals is crucial for [...] Read more.
The electrocardiogram (ECG) is one of the most extensively employed signals used in the diagnosis and prediction of cardiovascular diseases (CVDs). The ECG signals can capture the heart’s rhythmic irregularities, commonly known as arrhythmias. A careful study of ECG signals is crucial for precise diagnoses of patients’ acute and chronic heart conditions. In this study, we propose a two-dimensional (2-D) convolutional neural network (CNN) model for the classification of ECG signals into eight classes; namely, normal beat, premature ventricular contraction beat, paced beat, right bundle branch block beat, left bundle branch block beat, atrial premature contraction beat, ventricular flutter wave beat, and ventricular escape beat. The one-dimensional ECG time series signals are transformed into 2-D spectrograms through short-time Fourier transform. The 2-D CNN model consisting of four convolutional layers and four pooling layers is designed for extracting robust features from the input spectrograms. Our proposed methodology is evaluated on a publicly available MIT-BIH arrhythmia dataset. We achieved a state-of-the-art average classification accuracy of 99.11%, which is better than those of recently reported results in classifying similar types of arrhythmias. The performance is significant in other indices as well, including sensitivity and specificity, which indicates the success of the proposed method. Full article
(This article belongs to the Special Issue Interactive Deep Learning for Hyperspectral Images)
Show Figures

Graphical abstract

Open AccessArticle
Evaluation of Himawari-8/AHI, MERRA-2, and CAMS Aerosol Products over China
Remote Sens. 2020, 12(10), 1684; https://doi.org/10.3390/rs12101684 - 25 May 2020
Cited by 3 | Viewed by 818
Abstract
Reliable aerosol optical depth (AOD) data with high spatial and temporal resolutions are needed for research on air pollution in China. AOD products from the Advanced Himawari Imager (AHI) onboard the geostationary Himawari-8 satellite and reanalysis datasets make it possible to capture diurnal [...] Read more.
Reliable aerosol optical depth (AOD) data with high spatial and temporal resolutions are needed for research on air pollution in China. AOD products from the Advanced Himawari Imager (AHI) onboard the geostationary Himawari-8 satellite and reanalysis datasets make it possible to capture diurnal variations of aerosol loadings. However, due to the different retrieval methods, their applicability may vary with different space and time. Thus, in this study, taking the measured AOD at the Aerosol Robotic NETwork (AERONET) stations as the gold standard, the performance of the latest AHI hourly AOD product (i.e., L3 AOD) was evaluated and then compared with that of two reanalysis AOD datasets offered by Modern-Era Retrospective Analysis for Research and Applications, Version 2 (MERRA-2) and Copernicus Atmosphere Monitoring Service (CAMS), respectively, covering from July 2015 to December 2017 over China. For all the matchups, AHI AOD shows the highest robustness with a high correlation (R) of 0.82, low root-mean-square error (RMSE) of 0.23, and moderate mean absolute relative error (MARE) of 0.56. Although MERRA-2 and CAMS products both have lower R values (0.74, 0.72) and higher RMSE (0.28, 0.26), the former is slightly better than the latter. Accuracy of AOD products could be mainly affected by the pollution level and less affected by particle size distribution. Comparisons among these AOD products imply that AHI AOD is more reliable in regions with high pollution levels, such as central and eastern China, while in the northern and western part, MERRA-2 AOD seems more satisfying. The performance of all the three AOD products presents a significant diurnal variety, as indicated by the highest accuracy in the morning for AHI and at noon for reanalysis data. Moreover, due to various pollution distribution patterns and meteorological conditions, there are distinct seasonal characteristics in the performance of AOD products for different regions. Full article
Show Figures

Figure 1

Open AccessArticle
Coastal Wetland Mapping Using Ensemble Learning Algorithms: A Comparative Study of Bagging, Boosting and Stacking Techniques
Remote Sens. 2020, 12(10), 1683; https://doi.org/10.3390/rs12101683 - 25 May 2020
Cited by 6 | Viewed by 983
Abstract
Coastal wetlands are a critical component of the coastal landscape that are increasingly threatened by sea level rise and other human disturbance. Periodically mapping wetland distribution is crucial to coastal ecosystem management. Ensemble algorithms (EL), such as random forest (RF) and gradient boosting [...] Read more.
Coastal wetlands are a critical component of the coastal landscape that are increasingly threatened by sea level rise and other human disturbance. Periodically mapping wetland distribution is crucial to coastal ecosystem management. Ensemble algorithms (EL), such as random forest (RF) and gradient boosting machine (GBM) algorithms, are now commonly applied in the field of remote sensing. However, the performance and potential of other EL methods, such as extreme gradient boosting (XGBoost) and bagged trees, are rarely compared and tested for coastal wetland mapping. In this study, we applied the three most widely used EL techniques (i.e., bagging, boosting and stacking) to map wetland distribution in a highly modified coastal catchment, the Manning River Estuary, Australia. Our results demonstrated the advantages of using ensemble classifiers to accurately map wetland types in a coastal landscape. Enhanced bagging decision trees, i.e., classifiers with additional methods to increasing ensemble diversity such as RF and weighted subspace random forest, had comparably high predictive power. For the stacking method evaluated in this study, our results are inconclusive, and further comprehensive quantitative study is encouraged. Our findings also suggested that the ensemble methods were less effective at discriminating minority classes in comparison with more common classes. Finally, the variable importance results indicated that hydro-geomorphic factors, such as tidal depth and distance to water edge, were among the most influential variables across the top classifiers. However, vegetation indices derived from longer time series of remote sensing data that arrest the full features of land phenology are likely to improve wetland type separation in coastal areas. Full article
(This article belongs to the Section Environmental Remote Sensing)
Show Figures

Graphical abstract

Open AccessArticle
Feasibility Study Using UAV Aerial Photogrammetry for a Boundary Verification Survey of a Digitalized Cadastral Area in an Urban City of Taiwan
Remote Sens. 2020, 12(10), 1682; https://doi.org/10.3390/rs12101682 - 25 May 2020
Cited by 1 | Viewed by 1282
Abstract
In conducting land boundary verification surveys in digitalized cadastral areas in Taiwan, possible parcel points must be surveyed. These points are employed in the overlap analysis and map registration of possible parcel points and digitalized cadastral maps to identify the coordinates of parcel [...] Read more.
In conducting land boundary verification surveys in digitalized cadastral areas in Taiwan, possible parcel points must be surveyed. These points are employed in the overlap analysis and map registration of possible parcel points and digitalized cadastral maps to identify the coordinates of parcel points. Based on the computed horizontal distance and angle between control points and parcel points, parcels are staked out using ground surveys. Most studies survey possible parcel points using ground surveys with, for example, total stations. Compared with ground surveys, UAV (Unmanned Aerial Vehicle) aerial photogrammetry can provide more possible parcel points. Thus, an overlap analysis of digitalized cadastral maps, combined with the collection of possible parcel points, will be more comprehensive. In this study, a high-quality-medium format camera, with a 55 mm focal length, was carried on a rotary UAV to take images, with a 3 cm ground sampling distance (GSD), flying 300 m above the ground. The images were taken with an 80% end-lap and side-lap to increase the visibility of the terrain details for stereo-mapping. According to the test conducted in this study, UAV aerial photogrammetry can accurately provide supplementary control points and assist in the boundary verification of digitalized cadastral areas in Taiwan. Full article
Show Figures

Figure 1

Open AccessArticle
The Local Median Filtering Method for Correcting the Laser Return Intensity Information from Discrete Airborne Laser Scanning Data
Remote Sens. 2020, 12(10), 1681; https://doi.org/10.3390/rs12101681 - 24 May 2020
Viewed by 968
Abstract
Laser return intensity (LRI) information obtained from airborne laser scanning (ALS) data has been used to classify land cover types and to reveal canopy physiological features. However, the sensor-related and environmental parameters may introduce noise. In this study, we developed a local median [...] Read more.
Laser return intensity (LRI) information obtained from airborne laser scanning (ALS) data has been used to classify land cover types and to reveal canopy physiological features. However, the sensor-related and environmental parameters may introduce noise. In this study, we developed a local median filtering (LMF) method to point-by-point correct the LRI information. For each point, we deduced the reference variation range for its LRI. Then, we replaced the outliers of LRI with their local median values. To evaluate the LMF method, we assessed the discrepancy of LRI information from the same and diverse land cover types. Moreover, we used the corrected LRI to distinguish points from grass, road, and bare land, which were classified as ground type in ALS data. The results show that using the LMF method could increase the similarity of pointwise LRI from the same land cover type and the discrepancy of those from different kinds of targets. Using the LMF-corrected LRI could improve the overall classification accuracy of three land cover types by about 3% (all over 81%, κ ≥ 0.73, p < 0.05), compared to those using the original and range-normalized LRI. The sensor-related metrics brought more noise to the original LRI information than the environmental factors. Using the LMF method could effectively correct LRI information from historical ALS datasets. Full article
(This article belongs to the Special Issue 3D Modelling from Point Cloud: Algorithms and Methods)
Show Figures

Graphical abstract

Open AccessArticle
An Object-Based Bidirectional Method for Integrated Building Extraction and Change Detection between Multimodal Point Clouds
Remote Sens. 2020, 12(10), 1680; https://doi.org/10.3390/rs12101680 - 24 May 2020
Cited by 1 | Viewed by 1021
Abstract
Building extraction and change detection are two important tasks in the remote sensing domain. Change detection between airborne laser scanning data and photogrammetric data is vulnerable to dense matching errors, mis-alignment errors and data gaps. This paper proposes an unsupervised object-based method for [...] Read more.
Building extraction and change detection are two important tasks in the remote sensing domain. Change detection between airborne laser scanning data and photogrammetric data is vulnerable to dense matching errors, mis-alignment errors and data gaps. This paper proposes an unsupervised object-based method for integrated building extraction and change detection. Firstly, terrain, roofs and vegetation are extracted from the precise laser point cloud, based on “bottom-up” segmentation and clustering. Secondly, change detection is performed in an object-based bidirectional manner: Heightened buildings and demolished buildings are detected by taking the laser scanning data as reference, while newly-built buildings are detected by taking the dense matching data as reference. Experiments on two urban data sets demonstrate its effectiveness and robustness. The object-based change detection achieves a recall rate of 92.31% and a precision rate of 88.89% for the Rotterdam dataset; it achieves a recall rate of 85.71% and a precision rate of 100% for the Enschede dataset. It can not only extract unchanged building footprints, but also assign heightened or demolished labels to the changed buildings. Full article
Show Figures

Graphical abstract

Open AccessArticle
Simulating Multi-Directional Narrowband Reflectance of the Earth’s Surface Using ADAM (A Surface Reflectance Database for ESA’s Earth Observation Missions)
Remote Sens. 2020, 12(10), 1679; https://doi.org/10.3390/rs12101679 - 23 May 2020
Cited by 2 | Viewed by 1140
Abstract
The ADAM (A Surface Reflectance Database for ESA’s Earth Observation Missions) product (a climatological database coupled to its companion calculation toolkit) enables users to simulate realistic hyperspectral and directional global Earth surface reflectances (i.e., top-of-canopy/bottom-of-atmosphere) over the 240–4000 nm spectral range (at 1-nm [...] Read more.
The ADAM (A Surface Reflectance Database for ESA’s Earth Observation Missions) product (a climatological database coupled to its companion calculation toolkit) enables users to simulate realistic hyperspectral and directional global Earth surface reflectances (i.e., top-of-canopy/bottom-of-atmosphere) over the 240–4000 nm spectral range (at 1-nm resolution) and in any illumination/observation geometry, at 0.1° × 0.1° spatial resolution for a typical year. ADAM aims to support the preparation of optical Earth observation missions as well as the design of operational processing chains for the retrieval of atmospheric parameters by characterizing the expected surface reflectance, accounting for its anisotropy. Firstly, we describe (1) the methods used in the development of the gridded monthly ADAM climatologies (over land surfaces: monthly means of normalized reflectances derived from MODIS observations in seven spectral bands for the year 2005; over oceans: monthly means over the 1999–2009 period of chlorophyll content from SeaWiFS and of wind speed from SeaWinds), and (2) the underlying modeling approaches of ADAM toolkit to simulate the spectro-directional variations of the reflectance depending on the assigned surface type. Secondly, we evaluate ADAM simulation performances over land surfaces. A comparison against POLDER multi-spectral/multi-directional measurements for year 2008 shows reliable simulation results with root mean square differences below 0.027 and R2 values above 0.9 for most of the 14 land cover IGBP classes investigated, with no significant bias identified. Only for the “Snow and ice” class is the performance lower pointing to a limitation of climatological data to represent actual snow properties. An evaluation of the modeled reflectance in the specific backscatter direction against CALIPSO data reveals that ADAM tends to overestimate (underestimate) the so-called “hot-spot” by a factor of about 1.5 (1.5 to 2) for barren (vegetated) surfaces. Full article
(This article belongs to the Section AI Remote Sensing)
Show Figures

Graphical abstract

Open AccessArticle
Performance Evaluation of the Multiple Quantile Regression Model for Estimating Spatial Soil Moisture after Filtering Soil Moisture Outliers
Remote Sens. 2020, 12(10), 1678; https://doi.org/10.3390/rs12101678 - 23 May 2020
Viewed by 806
Abstract
The spatial distribution of soil moisture (SM) was estimated by a multiple quantile regression (MQR) model with Terra Moderate Resolution Imaging Spectroradiometer (MODIS) and filtered SM data from 2013 to 2015 in South Korea. For input data, observed precipitation and SM data were [...] Read more.
The spatial distribution of soil moisture (SM) was estimated by a multiple quantile regression (MQR) model with Terra Moderate Resolution Imaging Spectroradiometer (MODIS) and filtered SM data from 2013 to 2015 in South Korea. For input data, observed precipitation and SM data were collected from the Korea Meteorological Administration and various institutions monitoring SM. To improve the work of a previous study, prior to the estimation of SM, outlier detection using the isolation forest (IF) algorithm was applied to the observed SM data. The original observed SM data resulted in IF_SM data following outlier detection. This study obtained an average data removal rate of 20.1% at 58 stations. For various reasons, such as instrumentation, environment, and random errors, the original observed SM data contained approximately 20% uncertain data. After outlier detection, this study performed a regression analysis by estimating land surface temperature quantiles. The soil characteristics were considered through reclassification into four soil types (clay, loam, silt, and sand), and the five-day antecedent precipitation was considered in order to estimate the regression coefficient of the MQR model. For all soil types, the coefficient of determination (R2) and root mean square error (RMSE) values ranged from 0.25 to 0.77 and 1.86% to 12.21%, respectively. The MQR results showed a much better performance than that of the multiple linear regression (MLR) results, which yielded R2 and RMSE values of 0.20 to 0.66 and 1.08% to 7.23%, respectively. As a further illustration of improvement, the box plots of the MQR SM were closer to those of the observed SM than those of the MLR SM. This result indicates that the cumulative distribution functions (CDF) of MQR SM matched the CDF of the observed SM. Thus, the MQR algorithm with outlier detection can overcome the limitations of the MLR algorithm by reducing both the bias and variance. Full article
(This article belongs to the Special Issue Remote Sensing for Streamflow Simulation)
Show Figures

Graphical abstract

Open AccessArticle
Automatic Processing of Aerial LiDAR Data to Detect Vegetation Continuity in the Surroundings of Roads
Remote Sens. 2020, 12(10), 1677; https://doi.org/10.3390/rs12101677 - 23 May 2020
Cited by 1 | Viewed by 1212
Abstract
The optimization of forest management in the surroundings of roads is a necessary task in term of wildfire prevention and the mitigation of their effects. One of the reasons why a forest fire spreads is the presence of contiguous flammable material, both horizontally [...] Read more.
The optimization of forest management in the surroundings of roads is a necessary task in term of wildfire prevention and the mitigation of their effects. One of the reasons why a forest fire spreads is the presence of contiguous flammable material, both horizontally and vertically and, thus, vegetation management becomes essential in preventive actions. This work presents a methodology to detect the continuity of vegetation based on aerial Light Detection and Ranging (LiDAR) point clouds, in combination with point cloud processing techniques. Horizontal continuity is determined by calculating Cover Canopy Fraction (CCF). The results obtained show 50% of shrubs presence and 33% of trees presence in the selected case of study, with an error of 5.71%. Regarding vertical continuity, a forest structure composed of a single stratum represents 81% of the zone. In addition, the vegetation located in areas around the roads were mapped, taking into consideration the distances established in the applicable law. Analyses show that risky areas range from a total of 0.12 ha in a 2 m buffer and 0.48 ha in a 10 m buffer, representing a 2.4% and 9.5% of the total study area, respectively. Full article
(This article belongs to the Special Issue Point Cloud Processing and Analysis in Remote Sensing)
Show Figures

Graphical abstract

Open AccessArticle
Unseen Land Cover Classification from High-Resolution Orthophotos Using Integration of Zero-Shot Learning and Convolutional Neural Networks
Remote Sens. 2020, 12(10), 1676; https://doi.org/10.3390/rs12101676 - 23 May 2020
Cited by 9 | Viewed by 1108
Abstract
Zero-shot learning (ZSL) is an approach to classify objects unseen during the training phase and shown to be useful for real-world applications, especially when there is a lack of sufficient training data. Only a limited amount of works has been carried out on [...] Read more.
Zero-shot learning (ZSL) is an approach to classify objects unseen during the training phase and shown to be useful for real-world applications, especially when there is a lack of sufficient training data. Only a limited amount of works has been carried out on ZSL, especially in the field of remote sensing. This research investigates the use of a convolutional neural network (CNN) as a feature extraction and classification method for land cover mapping using high-resolution orthophotos. In the feature extraction phase, we used a CNN model with a single convolutional layer to extract discriminative features. In the second phase, we used class attributes learned from the Word2Vec model (pre-trained by Google News) to train a second CNN model that performed class signature prediction by using both the features extracted by the first CNN and class attributes during training and only the features during prediction. We trained and tested our models on datasets collected over two subareas in the Cameron Highlands (training dataset, first test dataset) and Ipoh (second test dataset) in Malaysia. Several experiments have been conducted on the feature extraction and classification models regarding the main parameters, such as the network’s layers and depth, number of filters, and the impact of Gaussian noise. As a result, the best models were selected using various accuracy metrics such as top-k categorical accuracy for k = [1,2,3], Recall, Precision, and F1-score. The best model for feature extraction achieved 0.953 F1-score, 0.941 precision, 0.882 recall for the training dataset and 0.904 F1-score, 0.869 precision, 0.949 recall for the first test dataset, and 0.898 F1-score, 0.870 precision, 0.838 recall for the second test dataset. The best model for classification achieved an average of 0.778 top-one, 0.890 top-two and 0.942 top-three accuracy, 0.798 F1-score, 0.766 recall and 0.838 precision for the first test dataset and 0.737 top-one, 0.906 top-two, 0.924 top-three, 0.729 F1-score, 0.676 recall and 0.790 precision for the second test dataset. The results demonstrated that the proposed ZSL is a promising tool for land cover mapping based on high-resolution photos. Full article
(This article belongs to the Special Issue Remote Sensing for Land Cover and Vegetation Mapping)
Show Figures

Graphical abstract

Open AccessArticle
Land Use Simulation of Guangzhou Based on Nighttime Light Data and Planning Policies
Remote Sens. 2020, 12(10), 1675; https://doi.org/10.3390/rs12101675 - 23 May 2020
Viewed by 934
Abstract
With the implementation processes of strategies such as Guangdong-Hong Kong-Macau Greater Bay Area’s coordinated development and “Belt and Road Initiative” initiative, the planning policies had produced a significant influence on land use distributions in Guangzhou. In this paper, we employ nighttime light (NTL) [...] Read more.
With the implementation processes of strategies such as Guangdong-Hong Kong-Macau Greater Bay Area’s coordinated development and “Belt and Road Initiative” initiative, the planning policies had produced a significant influence on land use distributions in Guangzhou. In this paper, we employ nighttime light (NTL) information as a proxy indicator of gross domestic product(GDP), and a future land use simulation model (FLUS) to simulate the land use patterns in Guangzhou from 2015 to 2018 and 2018 to 2035 by incorporating planning policies. The results show that: (1) the accuracy of simulation result from 2015 to 2018 based on National Polar-orbiting Partnership, Visible Infrared Imaging Radiometer Suite (NPP/VIIRS) is higher than that based on GDP; (2) by incorporating planning policies into the model can better identify the potential spatial distribution of urban land and make the simulated results more consistent with the actual urban land development trajectory. This study demonstrates that NTL is a suitable and feasible proxy indicator of GDP for the land use simulations, providing a scientific basis for the development of urban planning and construction policy. Full article
Show Figures

Graphical abstract

Open AccessArticle
Two-Path Network with Feedback Connections for Pan-Sharpening in Remote Sensing
Remote Sens. 2020, 12(10), 1674; https://doi.org/10.3390/rs12101674 - 23 May 2020
Cited by 2 | Viewed by 907
Abstract
High-resolution multi-spectral images are desired for applications in remote sensing. However, multi-spectral images can only be provided in low resolutions by optical remote sensing satellites. The technique of pan-sharpening wants to generate high-resolution multi-spectral (MS) images based on a panchromatic (PAN) image and [...] Read more.
High-resolution multi-spectral images are desired for applications in remote sensing. However, multi-spectral images can only be provided in low resolutions by optical remote sensing satellites. The technique of pan-sharpening wants to generate high-resolution multi-spectral (MS) images based on a panchromatic (PAN) image and the low-resolution counterpart. The conventional deep learning based pan-sharpening methods process the panchromatic and the low-resolution image in a feedforward manner where shallow layers fail to access useful information from deep layers. To make full use of the powerful deep features that have strong representation ability, we propose a two-path network with feedback connections, through which the deep features can be rerouted for refining the shallow features in a feedback manner. Specifically, we leverage the structure of a recurrent neural network to pass the feedback information. Besides, a power feature extraction block with multiple projection pairs is designed to handle the feedback information and to produce power deep features. Extensive experimental results show the effectiveness of our proposed method. Full article
Show Figures

Graphical abstract

Open AccessFeature PaperArticle
Diversity of Algorithm and Spectral Band Inputs Improves Landsat Monitoring of Forest Disturbance
Remote Sens. 2020, 12(10), 1673; https://doi.org/10.3390/rs12101673 - 23 May 2020
Cited by 5 | Viewed by 1225
Abstract
Disturbance monitoring is an important application of the Landsat times series, both to monitor forest dynamics and to support wise forest management at a variety of spatial and temporal scales. In the last decade, there has been an acceleration in the development of [...] Read more.
Disturbance monitoring is an important application of the Landsat times series, both to monitor forest dynamics and to support wise forest management at a variety of spatial and temporal scales. In the last decade, there has been an acceleration in the development of approaches designed to put the Landsat archive to use towards these causes. Forest disturbance mapping has moved from using individual change-detection algorithms, which implement a single set of decision rules that may not apply well to a range of scenarios, to compiling ensembles of such algorithms. One approach that has greatly reduced disturbance detection error has been to combine individual algorithm outputs in Random Forest (RF) ensembles trained with disturbance reference data, a process called stacking (or secondary classification). Previous research has demonstrated more robust and sensitive detection of disturbance using stacking with both multialgorithm ensembles and multispectral ensembles (which make use of a single algorithm applied to multiple spectral bands). In this paper, we examined several additional dimensions of this problem, including: (1) type of algorithm (represented by processes using one image per year vs. all historical images); (2) spectral band choice (including both the basic Landsat reflectance bands and several popular indices based on those bands); (3) number of algorithm/spectral-band combinations needed; and (4) the value of including both algorithm and spectral band diversity in the ensembles. We found that ensemble performance substantially improved per number of model inputs if those inputs were drawn from a diversity of both algorithms and spectral bands. The best models included inputs from both algorithms, using different variants of shortwave-infrared (SWIR) and near-infrared (NIR) reflectance. Further disturbance detection improvement may depend upon the development of algorithms which either interrogate SWIR and NIR in new ways or better highlight disturbance signals in the visible wavelengths. Full article
(This article belongs to the Special Issue Forest Canopy Disturbance Detection Using Satellite Remote Sensing)
Show Figures

Graphical abstract

Open AccessArticle
Satellite Imagery-Based Damage Assessment on Nineveh and Nebi Yunus Archaeological Site in Iraq
Remote Sens. 2020, 12(10), 1672; https://doi.org/10.3390/rs12101672 - 23 May 2020
Viewed by 1129
Abstract
During the last decades, archaeological site looting throughout Iraq has increased significantly up to a point where some of the most famous and relevant ancient Mesopotamian cities are currently threatened in their integrity. Several important archaeological monuments and artifacts have been destroyed, due [...] Read more.
During the last decades, archaeological site looting throughout Iraq has increased significantly up to a point where some of the most famous and relevant ancient Mesopotamian cities are currently threatened in their integrity. Several important archaeological monuments and artifacts have been destroyed, due to ISIL attacks and associated looting. Since 2016, the policies of the European Union have been increasingly harsh to condemn these atrocious acts of destruction. In such a scenario, the European Union Satellite Centre can be an invaluable instrument for the identification and assessment of the damage in areas occupied by ISIL. A detailed view of the damage suffered by the Nineveh and Nebi Yunus ancient sites, in Iraq, was assessed via visual inspection. The analysis was conducted considering the main events that occurred in the city of Mosul, between November 2013 and March 2018. More than 25 satellite images, new acquisitions and archived, supported by collateral data, allowed the detection and classification of the damage occurred over time. A description of the methodology and the classification of category and type of damage is presented. The results of the analysis confirm the dramatic levels of destruction that these two ancient sites have been suffering since 2013. The analysis reported in this paper is part of a wider study that the SatCen conducted in cooperation with the EU Counter-Terrorism Office and PRISM Office. The whole activity aimed at confirming to EU institutions the massive looting and trafficking operated in the area. The results have been provided to archaeologists in the field as well in support of local authorities who are trying to evaluate the current situation in the area. Full article
Show Figures

Figure 1

Open AccessArticle
Temporal Calibration of an Evaporation-Based Spatial Disaggregation Method of SMOS Soil Moisture Data
Remote Sens. 2020, 12(10), 1671; https://doi.org/10.3390/rs12101671 - 23 May 2020
Cited by 1 | Viewed by 847
Abstract
The resolution of current satellite surface soil moisture (SM) estimates is very low, of tens of kilometers, which proves to be insufficient for various agricultural and hydrological applications. Amongst the existing downscaling approaches of remotely sensed SM, DISPATCH (DISaggregation based on a Physical [...] Read more.
The resolution of current satellite surface soil moisture (SM) estimates is very low, of tens of kilometers, which proves to be insufficient for various agricultural and hydrological applications. Amongst the existing downscaling approaches of remotely sensed SM, DISPATCH (DISaggregation based on a Physical And Theoretical scale CHange) improves the resolution of SMOS (Soil Moisture and Ocean Salinity) soil moisture data using soil evaporative efficiency (SEE) estimates at high resolution (HR) and a SEE(SM) model implemented at low resolution (LR). Defined as the ratio of actual to potential soil evaporation, SEE can be derived from the remotely sensed land surface temperature (LST) and normalized difference vegetation index (NDVI). The current version of DISPATCH uses a linear SEE(SM) model. This study aims at improving the SEE(SM) model and testing different calibration strategies, to ultimately have more robust and better downscaled SM products. A nonlinear SEE(SM) model is introduced and its influence on the derived HR SM products is studied over a range of conditions. Each model, linear and nonlinear, is calibrated from remote sensing data on a daily and a multi-date basis. The approaches were tested over two mixed dry and irrigated areas in Catalonia, Spain, and over one dry area in Morocco. When using the linear model, better statistical results were generally obtained using a daily calibration (current version of DISPATCH), most notably over one Spanish site. However, the best results were systematically obtained for an annually calibrated nonlinear model, in terms of all metrics considered: correlation coefficient, slope of the linear regression, bias, unbiased root mean square error. In particular, when using the annually calibrated nonlinear SEE (SM) model, the temporal slope of the linear regression between disaggregated and in situ soil moisture increased to 1.16 and 0.75 for one Spanish site and for the Moroccan site (as opposed to 0.44 and 0.58, respectively, when using the linear model with a daily calibration). The temporal correlation coefficient increased to 0.47 and 0.54 over the Spanish sites (as opposed to 0.18 and 0.27, respectively, when using the linear model with a daily calibration). Those contrasted results indicate compensation effects between the model type and the calibration strategy. Taking into account studies that report the strong nonlinear behavior of the SEE with respect to SM, the introduction of the nonlinear SEE(SM) model in DISPATCH, combined with a multi-date calibration, is proven to perform significantly better under various conditions, leading to more robust disaggregated SM products. The SEE modeling based on the nonlinear SM model, with a multi-date calibration, could be integrated into the CATDS—Centre Aval de Traitement des Données SMOS as a future product, as well as into existing evapotranspiration models, which are based on a combination of thermal and microwave data. Full article
Show Figures

Figure 1

Open AccessLetter
BDD-Net: A General Protocol for Mapping Buildings Damaged by a Wide Range of Disasters Based on Satellite Imagery
Remote Sens. 2020, 12(10), 1670; https://doi.org/10.3390/rs12101670 - 22 May 2020
Cited by 1 | Viewed by 961
Abstract
The timely and accurate recognition of damage to buildings after destructive disasters is one of the most important post-event responses. Due to the complex and dangerous situations in affected areas, field surveys of post-disaster conditions are not always feasible. The use of satellite [...] Read more.
The timely and accurate recognition of damage to buildings after destructive disasters is one of the most important post-event responses. Due to the complex and dangerous situations in affected areas, field surveys of post-disaster conditions are not always feasible. The use of satellite imagery for disaster assessment can overcome this problem. However, the textural and contextual features of post-event satellite images vary with disaster types, which makes it difficult to use models that have been developed for a specific disaster type to detect damaged buildings following other types of disasters. Therefore, it is hard to use a single model to effectively and automatically recognize post-disaster building damage for a broad range of disaster types. Therefore, in this paper, we introduce a building damage detection network (BDD-Net) composed of a novel end-to-end remote sensing pixel-classification deep convolutional neural network. BDD-Net was developed to automatically classify every pixel of a post-disaster image into one of non-damaged building, damaged building, or background classes. Pre- and post-disaster images were provided as input for the network to increase semantic information, and a hybrid loss function that combines dice loss and focal loss was used to optimize the network. Publicly available data were utilized to train and test the model, which makes the presented method readily repeatable and comparable. The protocol was tested on images for five disaster types, namely flood, earthquake, volcanic eruption, hurricane, and wildfire. The results show that the proposed method is consistently effective for recognizing buildings damaged by different disasters and in different areas. Full article
Show Figures

Figure 1

Open AccessArticle
Comparison of Above-Water Seabird and TriOS Radiometers along an Atlantic Meridional Transect
Remote Sens. 2020, 12(10), 1669; https://doi.org/10.3390/rs12101669 - 22 May 2020
Cited by 4 | Viewed by 1046
Abstract
The Fiducial Reference Measurements for Satellite Ocean Color (FRM4SOC) project has carried out a range of activities to evaluate and improve the state-of-the-art in ocean color radiometry. This paper described the results from a ship-based intercomparison conducted on the Atlantic Meridional Transect 27 [...] Read more.
The Fiducial Reference Measurements for Satellite Ocean Color (FRM4SOC) project has carried out a range of activities to evaluate and improve the state-of-the-art in ocean color radiometry. This paper described the results from a ship-based intercomparison conducted on the Atlantic Meridional Transect 27 from 23rd September to 5th November 2017. Two different radiometric systems, TriOS-Radiation Measurement Sensor with Enhanced Spectral resolution (RAMSES) and Seabird-Hyperspectral Surface Acquisition System (HyperSAS), were compared and operated side-by-side over a wide range of Atlantic provinces and environmental conditions. Both systems were calibrated for traceability to SI (Système international) units at the same optical laboratory under uniform conditions before and after the field campaign. The in situ results and their accompanying uncertainties were evaluated using the same data handling protocols. The field data revealed variability in the responsivity between TRiOS and Seabird sensors, which is dependent on the ambient environmental and illumination conditions. The straylight effects for individual sensors were mostly within ±3%. A near infra-red (NIR) similarity correction changed the water-leaving reflectance (ρw) and water-leaving radiance (Lw) spectra significantly, bringing also a convergence in outliers. For improving the estimates of in situ uncertainty, it is recommended that additional characterization of radiometers and environmental ancillary measurements are undertaken. In general, the comparison of radiometric systems showed agreement within the evaluated uncertainty limits. Consistency of in situ results with the available Sentinel-3A Ocean and Land Color Instrument (OLCI) data in the range from (400…560) nm was also satisfactory (−8% < Mean Percentage Difference (MPD) < 15%) and showed good agreement in terms of the shape of the spectra and absolute values. Full article
(This article belongs to the Special Issue Fiducial Reference Measurements for Satellite Ocean Colour)
Show Figures

Graphical abstract

Open AccessArticle
Multi-Temporal Unmanned Aerial Vehicle Remote Sensing for Vegetable Mapping Using an Attention-Based Recurrent Convolutional Neural Network
Remote Sens. 2020, 12(10), 1668; https://doi.org/10.3390/rs12101668 - 22 May 2020
Cited by 7 | Viewed by 968
Abstract
Vegetable mapping from remote sensing imagery is important for precision agricultural activities such as automated pesticide spraying. Multi-temporal unmanned aerial vehicle (UAV) data has the merits of both very high spatial resolution and useful phenological information, which shows great potential for accurate vegetable [...] Read more.
Vegetable mapping from remote sensing imagery is important for precision agricultural activities such as automated pesticide spraying. Multi-temporal unmanned aerial vehicle (UAV) data has the merits of both very high spatial resolution and useful phenological information, which shows great potential for accurate vegetable classification, especially under complex and fragmented agricultural landscapes. In this study, an attention-based recurrent convolutional neural network (ARCNN) has been proposed for accurate vegetable mapping from multi-temporal UAV red-green-blue (RGB) imagery. The proposed model firstly utilizes a multi-scale deformable CNN to learn and extract rich spatial features from UAV data. Afterwards, the extracted features are fed into an attention-based recurrent neural network (RNN), from which the sequential dependency between multi-temporal features could be established. Finally, the aggregated spatial-temporal features are used to predict the vegetable category. Experimental results show that the proposed ARCNN yields a high performance with an overall accuracy of 92.80%. When compared with mono-temporal classification, the incorporation of multi-temporal UAV imagery could significantly boost the accuracy by 24.49% on average, which justifies the hypothesis that the low spectral resolution of RGB imagery could be compensated by the inclusion of multi-temporal observations. In addition, the attention-based RNN in this study outperforms other feature fusion methods such as feature-stacking. The deformable convolution operation also yields higher classification accuracy than that of a standard convolution unit. Results demonstrate that the ARCNN could provide an effective way for extracting and aggregating discriminative spatial-temporal features for vegetable mapping from multi-temporal UAV RGB imagery. Full article
(This article belongs to the Special Issue UAS-Remote Sensing Methods for Mapping, Monitoring and Modeling Crops)
Show Figures

Graphical abstract

Open AccessEditor’s ChoiceReview
Object Detection and Image Segmentation with Deep Learning on Earth Observation Data: A Review-Part I: Evolution and Recent Trends
Remote Sens. 2020, 12(10), 1667; https://doi.org/10.3390/rs12101667 - 22 May 2020
Cited by 19 | Viewed by 11800
Abstract
Deep learning (DL) has great influence on large parts of science and increasingly established itself as an adaptive method for new challenges in the field of Earth observation (EO). Nevertheless, the entry barriers for EO researchers are high due to the dense and [...] Read more.
Deep learning (DL) has great influence on large parts of science and increasingly established itself as an adaptive method for new challenges in the field of Earth observation (EO). Nevertheless, the entry barriers for EO researchers are high due to the dense and rapidly developing field mainly driven by advances in computer vision (CV). To lower the barriers for researchers in EO, this review gives an overview of the evolution of DL with a focus on image segmentation and object detection in convolutional neural networks (CNN). The survey starts in 2012, when a CNN set new standards in image recognition, and lasts until late 2019. Thereby, we highlight the connections between the most important CNN architectures and cornerstones coming from CV in order to alleviate the evaluation of modern DL models. Furthermore, we briefly outline the evolution of the most popular DL frameworks and provide a summary of datasets in EO. By discussing well performing DL architectures on these datasets as well as reflecting on advances made in CV and their impact on future research in EO, we narrow the gap between the reviewed, theoretical concepts from CV and practical application in EO. Full article
Show Figures

Graphical abstract

Open AccessArticle
Performance Modeling Framework for IoT-over-Satellite Using Shared Radio Spectrum
Remote Sens. 2020, 12(10), 1666; https://doi.org/10.3390/rs12101666 - 22 May 2020
Cited by 2 | Viewed by 1095
Abstract
Delivering Internet-of-Things (IoT) connectivity over satellite is a promising solution for applications in remote and sparsely populated areas. These applications range from smart agriculture, logistics, asset tracking to emergency services. Using a shared radio spectrum with terrestrial services will facilitate a cost-effective and [...] Read more.
Delivering Internet-of-Things (IoT) connectivity over satellite is a promising solution for applications in remote and sparsely populated areas. These applications range from smart agriculture, logistics, asset tracking to emergency services. Using a shared radio spectrum with terrestrial services will facilitate a cost-effective and rapid deployment of IoT-over-Satellite since it reduces the administrative and financial hurdles of leasing a dedicated segment of the spectrum. Although IoT-over-Satellite communication provides larger service coverage, the vast number of IoT devices also increase the interference in the satellite uplink channel, and it becomes a significant challenge for the reliable performance of the IoT-over-satellite. In this paper, we propose a framework for modeling the performance of IoT-over-Satellite access systems when sharing the radio spectrum with terrestrial networks. We take into consideration several important aspects, namely; satellite orbit, terrestrial IoT devices uplink interference, atmosphere and gas absorption, and the probability of line-of-sight. The performance of the overall system is presented in terms of the uplink signal-to-interference-plus-noise ratio (SINR), and thus the time-availability of the satellite link during a typical pass. We focus on low earth orbit satellites due to their potential use in IoT applications, where we evaluate the framework using actual parameters of satellites located in 300–800 km orbits. Furthermore, the paper presents a numercial model to obtain the most suitable antenna beamwidth that maximizes the link-availability of the satellite link by the simultaneous reduction in the terrestrial interference and the boosting of the underlying IoT signal of interest. Full article
(This article belongs to the Special Issue Satellite Communication)
Show Figures

Graphical abstract

Previous Issue
Next Issue
Back to TopTop