High Resolution Multispectral Data Research Articles

MLR-DBPFN: A Multi-Scale Low Rank Deep Back Projection Fusion Network for Anti-Noise Hyperspectral and Multispectral Image Fusion

Fusing low spatial resolution (LR) hyperspectral (HS) data and high spatial resolution (HR) multispectral (MS) data aims to obtain HR HS data. However, due to bad weather and the aging of sensor equipment, HS images usually contain a lot of noise, e.g., Gaussian noise, strip noise, and mixed noise, which would make the fused image have low quality. To solve this problem, we propose the multiscale low-rank deep back projection fusion network (MLR-DBPFN). First, HS and MS are superimposed, and multiscale spectral features of the stacked image are extracted through multiscale low-rank decomposition and convolution operation, which effectively removes noisy spectral features. Second, the upsampling and downsampling network mechanisms are used to extract the multiscale spatial features from each layer of spectral features. Finally, the multiscale spectral features and multiscale spatial features are combined for network training, and the weight of the noisy spectrum features is reduced through the network feedback mechanism, which suppresses the noisy spectrum and improves the noisy HS fusion performance. Experimental results on datasets of different noise demonstrate that MLR-DBPFN has superior spatial and spectral fidelity, comparative fusion quality, and robust antinoise performance compared with state-of-the-art methods.

Data-Driven Cloud Clustering via a Rotationally Invariant Autoencoder

Advanced satellite-born remote sensing instruments produce high-resolution multi-spectral data for much of the globe at a daily cadence. These datasets open up the possibility of improved understanding of cloud dynamics and feedback, which remain the biggest source of uncertainty in global climate model projections. As a step towards answering these questions, we describe an automated rotation-invariant cloud clustering (RICC) method that leverages deep learning autoencoder technology to organize cloud imagery within large datasets in an unsupervised fashion, free from assumptions about predefined classes. We describe both the design and implementation of this method and its evaluation, which uses a sequence of testing protocols to determine whether the resulting clusters: (1) are physically reasonable, (i.e., embody scientifically relevant distinctions); (2) capture information on spatial distributions, such as textures; (3) are cohesive and separable in latent space; and (4) are rotationally invariant, (i.e., insensitive to the orientation of an image). Results obtained when these evaluation protocols are applied to RICC outputs suggest that the resultant novel cloud clusters capture meaningful aspects of cloud physics, are appropriately spatially coherent, and are invariant to orientations of input images. Our results support the possibility of using an unsupervised data-driven approach for automated clustering and pattern discovery in cloud imagery.

Unmanned Aerial System-Based Multispectral Water Quality Monitoring in the Iberian Pyrite Belt (SW Spain)

Few studies have assessed mining-associated water pollution using spectral characteristics. We used high-resolution multispectral data acquired by unmanned aerial drones combined with in situ chemical data to assess water quality parameters in 12 relatively small water bodies located in the Tharsis complex, an abandoned mining area in the Iberian pyrite belt (SW Spain). The spectral bands of Micasense RedEdge-MX Dual and spectral band combinations were used jointly with physicochemical data to estimate water quality parameters and develop reliable empirical models using regression analysis. Physicochemical parameters including pH, ORP, EC, Al, Cu, Fe, Mn, S, Si, and Zn were estimated with high accuracy levels (0.81 < R2 < 0.99, 4 < RMSE% < 75, 0.01 < MAPE < 0.97). In contrast, the observed and modelled values for Ba, Ca, and Mg did not agree well (0.42 < R2 < 0.70). The best-fitted models were used to generate spatial distribution maps, providing information on water quality patterns. This study demonstrated that using empirical models to generate spatial distribution maps can be an effective and easy way to monitor acid mine drainage.

Damage assessment of chilli thrips using high resolution multispectral satellite data

&#x0D; Remote sensing technology offers an effective, rapid and reliable tool for assessing pest severity in vegetation. Ground based hyperspectral radiometry studies revealed significant difference in the reflectance spectra between healthy and thrip damaged vegetation. Space borne multispectral reflectance from Sentinel 2A satellite data of chilli thrip infested canopy has significant differences in red region (Band 4 – 664.6 nm), NIR region (Bands 5, 6, 7, 8 &amp; 8A having central wavelengths at 704.1, 740.5, 782.8 &amp; 832.8 nm, respectively) and SWIR region (Bands 11 &amp; 12 having central wavelengths at 1613.7 and 2202.4 nm). In this study, an attempt was made to discriminate healthy and pest affected chilli crop in the multispectral satellite imagery using several multispectral vegetation indices. Of these, land surface water index, LSWI (p=0.018) and normalized difference water index, NDWI (p=0.001) were found significant. These indices were used to classify chilli fields in the satellite imagery into severe, moderate and healthy classes. Superior performance of LSWI over NDWI with overall accuracy of 93.80 and Kappa Coefficient of 0.89 was observed. Moran's Index was used to study the spatial distribution of chilli thrips and observed strong clustering (I= 0.9073, p=0.0001).&#x0D;

Estimation of the Conifer-Broadleaf Ratio in Mixed Forests Based on Time-Series Data

Most natural forests are mixed forests, a mixed broadleaf-conifer forest is essentially a heterogeneously mixed pixel in remote sensing images. Satellite missions rely on modeling to acquire regional or global vegetation parameter products. However, these retrieval models often assume homogeneous conditions at the pixel level, resulting in a decrease in the inversion accuracy, which is an issue for heterogeneous forests. Therefore, information on the canopy composition of a mixed forest is the basis for accurately retrieving vegetation parameters using remote sensing. Medium and high spatial resolution multispectral time-series data are important sources for canopy conifer-broadleaf ratio estimation because these data have a high frequency and wide coverage. This paper highlights a successful method for estimating the conifer-broadleaf ratio in a mixed forest with diverse tree species and complex canopy structures. Experiments were conducted in the Purple Mountain, Nanjing, Jiangsu Province of China, where we collected leaf area index (LAI) time-series and forest sample plot inventory data. Based on the Invertible Forest Reflectance Model (INFORM), we simulated the normalized difference vegetation index (NDVI) time-series of different conifer-broadleaf ratios. A time-series similarity analysis was performed to determine the typical separable conifer-broadleaf ratios. Fifteen Gaofen-1 (GF-1) satellite images of 2015 were acquired. The conifer-broadleaf ratio estimation was based on the GF-1 NDVI time-series and semi-supervised k-means cluster method, which yielded a high overall accuracy of 83.75%. This study demonstrates the feasibility of accurately estimating separable conifer-broadleaf ratios using field measurement data and GF-1 time series in mixed broadleaf-conifer forests.

Bagging and Boosting Ensemble Classifiers for Classification of Multispectral, Hyperspectral and PolSAR Data: A Comparative Evaluation

In recent years, several powerful machine learning (ML) algorithms have been developed for image classification, especially those based on ensemble learning (EL). In particular, Extreme Gradient Boosting (XGBoost) and Light Gradient Boosting Machine (LightGBM) methods have attracted researchers’ attention in data science due to their superior results compared to other commonly used ML algorithms. Despite their popularity within the computer science community, they have not yet been well examined in detail in the field of Earth Observation (EO) for satellite image classification. As such, this study investigates the capability of different EL algorithms, generally known as bagging and boosting algorithms, including Adaptive Boosting (AdaBoost), Gradient Boosting Machine (GBM), XGBoost, LightGBM, and Random Forest (RF), for the classification of Remote Sensing (RS) data. In particular, different classification scenarios were designed to compare the performance of these algorithms on three different types of RS data, namely high-resolution multispectral, hyperspectral, and Polarimetric Synthetic Aperture Radar (PolSAR) data. Moreover, the Decision Tree (DT) single classifier, as a base classifier, is considered to evaluate the classification’s accuracy. The experimental results demonstrated that the RF and XGBoost methods for the multispectral image, the LightGBM and XGBoost methods for hyperspectral data, and the XGBoost and RF algorithms for PolSAR data produced higher classification accuracies compared to other ML techniques. This demonstrates the great capability of the XGBoost method for the classification of different types of RS data.

Investigating a Selection of Methods for the Prediction of Total Soluble Solids Among Wine Grape Quality Characteristics Using Normalized Difference Vegetation Index Data From Proximal and Remote Sensing.

The most common method for determining wine grape quality characteristics is to perform sample-based laboratory analysis, which can be time-consuming and expensive. In this article, we investigate an alternative approach to predict wine grape quality characteristics by combining machine learning techniques and normalized difference vegetation index (NDVI) data collected at different growth stages with non-destructive methods, such as proximal and remote sensing, that are currently used in precision viticulture (PV). The study involved several sets of high-resolution multispectral data derived from four sources, including two vehicle-mounted crop reflectance sensors, unmanned aerial vehicle (UAV)-acquired data, and Sentinel-2 (S2) archived imagery to estimate grapevine canopy properties at different growth stages. Several data pre-processing techniques were employed, including data quality assessment, data interpolation onto a 100-cell grid (10 × 20 m), and data normalization. By calculating Pearson’s correlation matrix between all variables, initial descriptive statistical analysis was carried out to investigate the relationships between NDVI data from all proximal and remote sensors and the grape quality characteristics in all growth stages. The transformed dataset was then ready and applied to statistical and machine learning algorithms, firstly trained on the data distribution available and then validated and tested, using linear and nonlinear regression models, including ordinary least square (OLS), Theil–Sen, and the Huber regression models and Ensemble Methods based on Decision Trees. Proximal sensors performed better in wine grapes quality parameters prediction in the early season, while remote sensors during later growth stages. The strongest correlations with the sugar content were observed for NDVI data collected with the UAV, Spectrosense+GPS (SS), and the CropCircle (CC), during Berries pea-sized and the Veraison stage, mid-late season with full canopy growth, for both years. UAV and SS data proved to be more accurate in predicting the sugars out of all wine grape quality characteristics, especially during a mid-late season with full canopy growth, in Berries pea-sized and the Veraison growth stages. The best-fitted regressions presented a maximum coefficient of determination (R2) of 0.61.

Monitoring Sea Surface Currents Based on China GF-4 Multispectral Data

The seawater tracing index that characterizes the sea surface feature can be calculated using the optimal band combination of China GF-4 high-resolution multispectral data, and the distribution of sea surface currents can be retrieved from the sequential band combination data. In this paper, the methods of monitoring sea surface currents are presented in detail by processing multispectral remote sensing images generated by China GF-4 Earth Observation satellite. The sea surface currents obtained from GF-4 imagery data are compared with the sea surface currents derived from FES2014 global ocean tide model, providing a validation of the proposed methods.

Shallow Water Bathymetry Mapping of Xinji Island Based on Multispectral Satellite Image using Deep Learning

Nearshore bathymetry is a basic parameter of the ocean, which is crucial to the research and management of coastal zones. Previous studies have demonstrated that remote sensing techniques can be employed in estimating bathymetric information. In this paper, we propose a deep belief network with data perturbation (DBN-DP) algorithm for shallow water depth inversion from high resolution multispectral data, and applying it in Xinji Island of Malacca Strait and Yongxing Island in China. Results show that the DBN-DP method can produce more accurate water depth estimations than other traditional methods particularly for deeper water, which reaches 1.2 m of mean absolute error (MAE) and 12.8% of mean relative error (MRE) in Xinji Island. Most of the estimated bathymetry meet the category of zone of confidence C level defined by the International Hydrographic Organization. These findings are encouraging for employing deep learning in bathymetry, which may become a novel approach for bathymetric inversion in the future.

Monitoring Drought using Multispectral Remote Sensing – A Case Study

The significance of remote sensing observations in monitoring environmental phenomena is a well-established fact. High resolution satellite images acquired using multispectral sensors contribute greatly in this aspect. In the current study, an illustration of this factor is presented using high resolution multispectral data to analyse drought conditions over Solapur, India. Here, Sentinel-2 data is utilized to estimate indices like Normalized Difference Vegetation Index (NDVI), Normalized Difference Water Index (NDWI) and Vegetation condition index (VCI), which help us to infer the drought situation in the region of interest.

Evaluation of Features Derived from High-Resolution Multispectral Imagery and LiDAR Data for Object-Based Support Vector Machine Classification of Tree Species

Remote sensing can play a key role in understanding the make-up of urban forests. This study analyzes how high-resolution Geoeye-1 multispectral imagery and LiDAR point clouds allow for improved classification of urban tree species using object-based and support vector machine classification (SVM). Five common urban trees are classified: Acer platanoides; Acer platanoides ‘Schwedleri’; Picea pungens; Gleditsia triacanthos; and Tilia cordata. Numerous features are used for classification: index derived from imagery reflectance; texture of imagery; LiDAR height and intensity; and a LiDAR-generated normalized digital surface model. Classification is performed to evaluate the contribution of individual features, groups of features, and the combination of features from both imagery and LiDAR data. Classification results in an overall accuracy of 85.08% when features from both data sources are combined, compared with 77.73% when using only LiDAR features, and 71.85% when using only imagery features.

A Classified Adversarial Network for Multi-Spectral Remote Sensing Image Change Detection

Adversarial training has demonstrated advanced capabilities for generating image models. In this paper, we propose a deep neural network, named a classified adversarial network (CAN), for multi-spectral image change detection. This network is based on generative adversarial networks (GANs). The generator captures the distribution of the bitemporal multi-spectral image data and transforms it into change detection results, and these change detection results (as the fake data) are input into the discriminator to train the discriminator. The results obtained by pre-classification are also input into the discriminator as the real data. The adversarial training can facilitate the generator learning the transformation from a bitemporal image to a change map. When the generator is trained well, the generator has the ability to generate the final result. The bitemporal multi-spectral images are input into the generator, and then the final change detection results are obtained from the generator. The proposed method is completely unsupervised, and we only need to input the preprocessed data that were obtained from the pre-classification and training sample selection. Through adversarial training, the generator can better learn the relationship between the bitemporal multi-spectral image data and the corresponding labels. Finally, the well-trained generator can be applied to process the raw bitemporal multi-spectral images to obtain the final change map (CM). The effectiveness and robustness of the proposed method were verified by the experimental results on the real high-resolution multi-spectral image data sets.

Remote characterization of photosynthetic communities in the Fryxell basin of Taylor Valley, Antarctica

AbstractWe investigate the spatial distribution, spectral properties and temporal variability of primary producers (e.g. communities of microbial mats and mosses) throughout the Fryxell basin of Taylor Valley, Antarctica, using high-resolution multispectral remote-sensing data. Our results suggest that photosynthetic communities can be readily detected throughout the Fryxell basin based on their unique near-infrared spectral signatures. Observed intra- and inter-annual variability in spectral signatures are consistent with short-term variations in mat distribution, hydration and photosynthetic activity. Spectral unmixing is also implemented in order to estimate mat abundance, with the most densely vegetated regions observed from orbit correlating spatially with some of the most productive regions of the Fryxell basin. Our work establishes remote sensing as a valuable tool in the study of these ecological communities in the McMurdo Dry Valleys and demonstrates how future scientific investigations and the management of specially protected areas could benefit from these tools and techniques.

Fusing China GF-5 Hyperspectral Data with GF-1, GF-2 and Sentinel-2A Multispectral Data: Which Methods Should Be Used?

The China GaoFen-5 (GF-5) satellite sensor, which was launched in 2018, collects hyperspectral data with 330 spectral bands, a 30 m spatial resolution, and 60 km swath width. Its competitive advantages compared to other on-orbit or planned sensors are its number of bands, spectral resolution, and swath width. Unfortunately, its applications may be undermined by its relatively low spatial resolution. Therefore, the data fusion of GF-5 with high spatial resolution multispectral data is required to further enhance its spatial resolution while preserving its spectral fidelity. This paper conducted a comprehensive evaluation study of fusing GF-5 hyperspectral data with three typical multispectral data sources (i.e., GF-1, GF-2 and Sentinel-2A (S2A)), based on quantitative metrics, classification accuracy, and computational efficiency. Datasets on three study areas of China were utilized to design numerous experiments, and the performances of nine state-of-the-art fusion methods were compared. Experimental results show that LANARAS (this method was proposed by lanaras et al.), Adaptive Gram–Schmidt (GSA), and modulation transfer function (MTF)-generalized Laplacian pyramid (GLP) methods are more suitable for fusing GF-5 with GF-1 data, MTF-GLP and GSA methods are recommended for fusing GF-5 with GF-2 data, and GSA and smoothing filtered-based intensity modulation (SFIM) can be used to fuse GF-5 with S2A data.

Mapping Vegetation at Species Level with High-Resolution Multispectral and Lidar Data Over a Large Spatial Area: A Case Study with Kudzu

Mapping vegetation species is critical to facilitate related quantitative assessment, and mapping invasive plants is important to enhance monitoring and management activities. Integrating high-resolution multispectral remote-sensing (RS) images and lidar (light detection and ranging) point clouds can provide robust features for vegetation mapping. However, using multiple sources of high-resolution RS data for vegetation mapping on a large spatial scale can be both computationally and sampling intensive. Here, we designed a two-step classification workflow to potentially decrease computational cost and sampling effort and to increase classification accuracy by integrating multispectral and lidar data in order to derive spectral, textural, and structural features for mapping target vegetation species. We used this workflow to classify kudzu, an aggressive invasive vine, in the entire Knox County (1362 km2) of Tennessee (U.S.). Object-based image analysis was conducted in the workflow. The first-step classification used 320 kudzu samples and extensive, coarsely labeled samples (based on national land cover) to generate an overprediction map of kudzu using random forest (RF). For the second step, 350 samples were randomly extracted from the overpredicted kudzu and labeled manually for the final prediction using RF and support vector machine (SVM). Computationally intensive features were only used for the second-step classification. SVM had constantly better accuracy than RF, and the producer’s accuracy, user’s accuracy, and Kappa for the SVM model on kudzu were 0.94, 0.96, and 0.90, respectively. SVM predicted 1010 kudzu patches covering 1.29 km2 in Knox County. We found the sample size of kudzu used for algorithm training impacted the accuracy and number of kudzu predicted. The proposed workflow could also improve sampling efficiency and specificity. Our workflow had much higher accuracy than the traditional method conducted in this research, and could be easily implemented to map kudzu in other regions as well as map other vegetation species.

Sea Ice Automatic Extraction in the Liaodong Bay from Sentinel-2 Imagery Using Convolutional Neural Networks

Sea ice classification is one of the important tasks of sea ice monitoring. Accurate extraction of sea ice types is of great significance on sea ice conditions assessment, smooth navigation and safty marine operations. Sentinel-2 is an optical satellite launched by the European Space Agency. High spatial resolution and wide range imaging provide powerful support for sea ice monitoring. However, traditional supervised classification method is difficult to achieve fine results for small sample features. In order to solve the problem, this paper proposed a sea ice extraction method based on deep learning and it was applied to Liaodong Bay in Bohai Sea, China. The convolutional neural network was used to extract and classify the feature of the image from Sentinel-2. The results showed that the overall accuracy of the algorithm was 85.79% which presented a significant improvement compared with the tranditional algorithms, such as minimum distance method, maximum likelihood method, Mahalanobis distance method, and support vector machine method. The method proposed in this paper, which combines convolutional neural networks and high-resolution multispectral data, provides a new idea for remote sensing monitoring of sea ice.

Remote spectral methods for detecting stress coniferous

The article presents investigation of the possibility of drying coniferous forest areas detecting by multispectral satellite data in the visible and NIR spectral range with low spatial resolution, obtained by the imaging systems of three satellites - the Belarusian spacecraft (BS), Landsat 8 and Sentinel 2. A forest area in the south of Belarus was considered as a test site. High-resolution multispectral airborne data and, in part, ground measurements were used as reference ground data by which training samples were formed. Most of the known classical methods of supervised classification have been tested, the maximum likelihood method turned out to be the best for this task. In order to improve the accuracy of identifying the drying areas of coniferous forests on multispectral images, parametric transformations of images in the spectral space are proposed, which should lead to an increase in initial small spectral differences. The methodological issues of assessing the accuracy of the satellite images classification are considered using the result of the classification of airborne image with high spatial resolution as a ground truth image. The assessment of the classification accuracy, both visually and using the obtained confusion matrices, allows us to conclude that the images of the BS, Landsat 8 and Sentinel 2 can be used to detect drying area of coniferous forests as well as the expediency of carrying out the proposed transformations of the original images.

Combining Unmanned Aerial Systems and Satellite Data to Monitor Phenological Changes in Tropical Forests: A Case Study from Costa Rica

The migration of vegetation under the influence of climate change is of great interest to ecologists, but can be difficult to quantify—especially in less accessible landscapes. Monitoring land cover change through remote sensing has become the best solution, especially with the use of unmanned aerial systems (UASs; drones) as low-cost remote sensing platforms are able to collect data at high spatial and spectral resolutions. Unfortunately, in the context of climate change studies, the lack of comparative UAS data sets over decadal timescales has been limiting. Here, we describe a technique for the integration of historical, low spatial resolution satellite-based Normalized Difference Vegetation Index (NDVI) data with short-term high-resolution multispectral UAS data to track the vegetation changes in a Costa Rican rainforest over a 33-year time frame. The study reveals the transition of a mixed forest from strongly deciduous to weakly deciduous phenology in the Hacienda Barú National Wildlife Refuge (HBNWR), southwestern Costa Rica. This case study presents an approach for researchers and forest managers to study and track vegetation changes over time in locations that lack detailed historical vegetation data. Vegetation migration due to climate change is not well documented and difficult to monitor, especially in remote or inaccessible locations. This case study presents researchers, students, and forest managers an approach for leveraging freely available satellite imagery and UASs to track these changes over time.

Airborne LiDAR and high resolution multispectral data integration in Eucalyptus tree species mapping in an Australian farmscape

Rapid decline and death of rural Eucalypts trees of all ages and species have been reported in the farmscapes of regional Australia due to various environmental and farming management related factors. The identification of existing farm tree species is important for long term management strategies to provide ecosystem stability in the region. This study explored the feasibility of structural attributes of LiDAR and spectral and spatial characteristics of high resolution remote sensing data to identify and map Eucalyptus tree species. An object based image segmentation and rule-based classification algorithm were developed to delineate tree boundaries and species classification. The integration of two datasets improved the classification accuracy (65%) against their separate classification (52% and 41%, respectively). The identification of tree species will help in getting first-hand information on existing farm trees, which may be used in assessing tree condition in time series related to management practices and complex dieback problem.

Analysis of Suspended Sediment Variability in a Large Highly Turbid Estuary Using a 5‐Year‐Long Remotely Sensed Data Archive at High Resolution

AbstractSpatial variability of surface suspended particulate matter (SPM) concentration in the Gironde estuary and their relationships with environmental forcing are investigated through high spatial resolution multispectral data collected from July 2013 to August 2018 by the Operational Land Imager (Landsat‐8/OLI) and MultiSpectral Instrument (Sentinel‐2/MSI). A principal component analysis using the T‐mode orientation is applied to the discontinuous multiannual time series composed of 41 remotely sensed images. The three first principal components (PC1, PC2, and PC3) explain 65.7% of the total variance. The SPM distribution associated with PC1 and PC2 exhibits a privileged along‐estuary direction of the oscillation modes, while the spatial patterns of PC3 are clearly dominated by lateral oscillations opposing channels and shoals. The main environmental factors affecting the SPM distribution are identified by the analysis of their temporal patterns. The tidal range and the daily river discharge control the spatial patterns of PC1 and PC2, while the tidal cycles and the wind speed are significantly correlated with PC3. Furthermore, the analysis in the along‐estuary and lateral directions shows marked longitudinal and transverse SPM gradients and a strong control of bathymetry on the SPM spatial distribution. For the first time, we highlight that the maxima of surface residual turbidity are located on the shoals regardless of the environmental (tidal and hydrological) conditions. Compared to previous studies, usually based on single‐point in‐water column measurements, these results provide a novel and complementary description of the spatial variability of SPM. They are useful to validate sediment transport numerical models, but also may improve our understanding of suspended sediment dynamics in estuarine systems governed by an estuarine turbidity maximum.