This PDF file contains the editorial “Special Section Guest Editorial: Feature and Deep Learning in Remote Sensing Application” for JARS Vol. 11 Issue 4.

Object tracking is a core subject in computer vision and has significant meaning in both theory and practice. We propose a tracking method in which a robust discriminative classifier is built based on both object and context information. In this method, we consider multiple frames of local invariant features on and around the object and construct the object template and context template. To overcome the limitation of the invariant representations, we also design a nonparametric learning algorithm using transitive matching perspective transformation. This learning algorithm can keep adding object appearance and can avoid improper updating when occlusions appear. We also analyze the asymptotic stability of our method and prove its drift-free capability in long-term tracking. Extensive experiments using challenging publicly available video sequences that cover most of the critical conditions in tracking demonstrate the enhanced strength and robustness of our method.

Existing methods to detect vehicle tracks in coherent change detection images, a product of combining two synthetic aperture radar images taken at different times of the same scene, rely on simple and fast models to label track pixels. These models, however, are unable to capture natural track features, such as continuity and parallelism. More powerful but computationally expensive models can be used in offline settings. We present an approach that uses dilated convolutional networks consisting of a series of 3×3 convolutions to segment vehicle tracks. The design of our networks considers the fact that remote sensing applications tend to operate in low power and have limited training data. As a result, we aim for small and efficient networks that can be trained end-to-end to learn natural track features entirely from limited training data. We demonstrate that our six-layer network, trained on just 90 images, is computationally efficient and improves the F-score on a standard dataset to 0.992, up from 0.959 obtained by the current state-of-the-art method.

Classification of hyperspectral remote sensing imagery is one of the most popular topics because of its intrinsic potential to gather spectral signatures of materials and provides distinct abilities to object detection and recognition. In the last decade, an enormous number of methods were suggested to classify hyperspectral remote sensing data using spectral features, though some are not using all information and lead to poor classification accuracy; on the other hand, the exploration of deep features is recently considered a lot and has turned into a research hot spot in the geoscience and remote sensing research community to enhance classification accuracy. A deep learning architecture is proposed to classify hyperspectral remote sensing imagery by joint utilization of spectral–spatial information. A stacked sparse autoencoder provides unsupervised feature learning to extract high-level feature representations of joint spectral–spatial information; then, a soft classifier is employed to train high-level features and to fine-tune the deep learning architecture. Comparative experiments are performed on two widely used hyperspectral remote sensing data (Salinas and PaviaU) and a coarse resolution hyperspectral data in the long-wave infrared range. The obtained results indicate the superiority of the proposed spectral–spatial deep learning architecture against the conventional classification methods.

Hyperspectral anomaly detection (AD) is an important technique of unsupervised target detection and has significance in real situations. Due to the high dimensionality of hyperspectral data, AD will be influenced by noise, nonlinear correlation of band, or other factors that lead to the decline of detection accuracy. To overcome this problem, a method of hyperspectral AD based on stacked denoising autoencoders (AE) (HADSDA) is proposed. Simultaneously, two different feature detection models, spectral feature (SF) and fused feature by clustering (FFC), are constructed to verify the effectiveness of the proposed algorithm. The SF detection model uses the SF of each pixel. The FFC detection model uses a similar set of pixels constructed by clustering and then fuses the set of pixels by the stacked denoising autoencoders algorithm (SDA). The SDA is an algorithm that can automatically learn nonlinear deep features of the image. Compared with other linear or nonlinear feature extraction methods, the detection result of the proposed algorithm is greatly improved. Experiment results show that the proposed algorithm is an excellent feature learning method and can achieve higher detection performance.

Deep convolutional neural networks (CNNs) have shown outstanding performance in object recognition from natural images. In contrast, object recognition from remote sensing images is more challenging, due to the complex background and inadequate data for training a deep network with a huge number of parameters. We propose a unified deep CNN, called DeepPlane, to simultaneously detect the position and classify the category of aircraft in remote sensing images. This model consists of two correlative deep networks: the first one is designed to generate object proposals as well as feature maps and the second one is cascaded upon the first one to perform classification and box regression in one shot. The “inception module” is introduced to tackle the insufficient training data problem that is one of the most challenging obstacles of detection in remote sensing images. Extensive experiments demonstrate the efficiency of the proposed DeepPlane model. Specifically, DeepPlane could model detection and classification jointly and achieves 91.9% mAP in six categories of aircraft, which advances the state-of-the-art, sometimes considerably, for both tasks.

Recent advances in remote sensing technology have made multisensor data available for the same area, and it is well-known that remote sensing data processing and analysis often benefit from multisource data fusion. Specifically, low spatial resolution of hyperspectral imagery (HSI) degrades the quality of the subsequent classification task while using visible (VIS) images with high spatial resolution enables high-fidelity spatial analysis. A collaborative classification framework is proposed to fuse HSI and VIS images for finer classification. First, the convolutional neural network model is employed to extract deep spectral features for HSI classification. Second, effective binarized statistical image features are learned as contextual basis vectors for the high-resolution VIS image, followed by a classifier. The proposed approach employs diversified data in a decision fusion, leading to an integration of the rich spectral information, spatial information, and statistical representation information. In particular, the proposed approach eliminates the potential problems of the curse of dimensionality and excessive computation time. The experiments evaluated on two standard data sets demonstrate better classification performance offered by this framework.

Automated classification of images across image archives requires reducing the semantic gap between high-level features perceived by humans and low-level features encoded in images. Due to rapidly growing image archives in the Earth science domain, it is critical to automatically classify images for efficient sorting and discovery. In particular, classifying images based on the presence of Earth science phenomena allows users to perform climatology studies and investigate case studies. We present applications of deep learning-based classification of Earth science images.

In recent years, deep learning (DL), a rebranding of neural networks (NNs), has risen to the top in numerous areas, namely computer vision (CV), speech recognition, and natural language processing. Whereas remote sensing (RS) possesses a number of unique challenges, primarily related to sensors and applications, inevitably RS draws from many of the same theories as CV, e.g., statistics, fusion, and machine learning, to name a few. This means that the RS community should not only be aware of advancements such as DL, but also be leading researchers in this area. Herein, we provide the most comprehensive survey of state-of-the-art RS DL research. We also review recent new developments in the DL field that can be used in DL for RS. Namely, we focus on theories, tools, and challenges for the RS community. Specifically, we focus on unsolved challenges and opportunities as they relate to (i) inadequate data sets, (ii) human-understandable solutions for modeling physical phenomena, (iii) big data, (iv) nontraditional heterogeneous data sources, (v) DL architectures and learning algorithms for spectral, spatial, and temporal data, (vi) transfer learning, (vii) an improved theoretical understanding of DL systems, (viii) high barriers to entry, and (ix) training and optimizing the DL.

Ocean fronts have been a subject of study for many years, a variety of methods and algorithms have been proposed to address the problem of ocean fronts. However, all these existing ocean front recognition methods are built upon human expertise in defining the front based on subjective thresholds of relevant physical variables. This paper proposes a deep learning approach for ocean front recognition that is able to automatically recognize the front. We first investigated four existing deep architectures, i.e., AlexNet, CaffeNet, GoogLeNet, and VGGNet, for the ocean front recognition task using remote sensing (RS) data. We then propose a deep network with fewer layers compared to existing architecture for the front recognition task. This network has a total of five learnable layers. In addition, we extended the proposed network to recognize and classify the front into strong and weak ones. We evaluated and analyzed the proposed network with two strategies of exploiting the deep model: full-training and fine-tuning. Experiments are conducted on three different RS image datasets, which have different properties. Experimental results show that our model can produce accurate recognition results.

Automatic ship detection in optical remote sensing images has attracted wide attention for its broad applications. Major challenges for this task include the interference of cloud, wave, wake, and the high computational expenses. We propose a fast and robust ship detection algorithm to solve these issues. The framework for ship detection is designed based on deep convolutional neural networks (CNNs), which provide the accurate locations of ship targets in an efficient way. First, the deep CNN is designed to extract features. Then, a region proposal network (RPN) is applied to discriminate ship targets and regress the detection bounding boxes, in which the anchors are designed by intrinsic shape of ship targets. Experimental results on numerous panchromatic images demonstrate that, in comparison with other state-of-the-art ship detection methods, our method is more efficient and achieves higher detection accuracy and more precise bounding boxes in different complex backgrounds.

We investigate the effectiveness of deep neural network for cross-domain classification of remote sensing images in this paper. In the network, class centroid alignment is utilized as a domain adaptation strategy, making the network able to transfer knowledge from the source domain to target domain on a per-class basis. Since predicted labels of target data should be used to estimate the centroid of each class, we use overall centroid alignment as a coarse domain adaptation method to improve the estimation accuracy. In addition, rectified linear unit is used as the activation function to produce sparse features, which may improve the separation capability. The proposed network can provide both aligned features and an adaptive classifier, as well as obtain label-free classification of target domain data. The experimental results using Hyperion, NCALM, and WorldView-2 remote sensing images demonstrated the effectiveness of the proposed approach.

Deep convolutional neural networks (CNNs) have been widely used to obtain high-level representation in various computer vision tasks. However, for remote scene classification, there are not sufficient images to train a very deep CNN from scratch. From two viewpoints of generalization power, we propose two promising kinds of deep CNNs for remote scenes and try to find whether deep CNNs need to be deep for remote scene classification. First, we transfer successful pretrained deep CNNs to remote scenes based on the theory that depth of CNNs brings the generalization power by learning available hypothesis for finite data samples. Second, according to the opposite viewpoint that generalization power of deep CNNs comes from massive memorization and shallow CNNs with enough neural nodes have perfect finite sample expressivity, we design a lightweight deep CNN (LDCNN) for remote scene classification. With five well-known pretrained deep CNNs, experimental results on two independent remote-sensing datasets demonstrate that transferred deep CNNs can achieve state-of-the-art results in an unsupervised setting. However, because of its shallow architecture, LDCNN cannot obtain satisfactory performance, regardless of whether in an unsupervised, semisupervised, or supervised setting. CNNs really need depth to obtain general features for remote scenes. This paper also provides baseline for applying deep CNNs to other remote sensing tasks.

We evaluated how deep convolutional neural networks (DCNN) could assist in the labor-intensive process of human visual searches for objects of interest in high-resolution imagery over large areas of the Earth’s surface. Various DCNN were trained and tested using fewer than 100 positive training examples (China only) from a worldwide surface-to-air-missile (SAM) site dataset. A ResNet-101 DCNN achieved a 98.2% average accuracy for the China SAM site data. The ResNet-101 DCNN was used to process ∼19.6  M image chips over a large study area in southeastern China. DCNN chip detections (∼9300) were postprocessed with a spatial clustering algorithm to produce a ranked list of ∼2100 candidate SAM site locations. The combination of DCNN processing and spatial clustering effectively reduced the search area by ∼660X (0.15% of the DCNN-processed land area). An efficient web interface was used to facilitate a rapid serial human review of the candidate SAM sites in the China study area. Four novice imagery analysts with no prior imagery analysis experience were able to complete a DCNN-assisted SAM site search in an average time of ∼42  min. This search was ∼81X faster than a traditional visual search over an equivalent land area of ∼88,640  km2 while achieving nearly identical statistical accuracy (∼90% F1).

This paper proposes a model of dual-channel convolutional neural network (CNN) that is designed for change detection in SAR images, in an effort to acquire higher detection accuracy and lower misclassification rate. This network model contains two parallel CNN channels, which can extract deep features from two multitemporal SAR images. For comparison and validation, the proposed method is tested along with other change detection algorithms on both simulated SAR images and real-world SAR images captured by different sensors. The experimental results demonstrate that the presented method outperforms the state-of-the-art techniques by a considerable margin.

Feature extraction has always been a difficult problem in the classification performance of synthetic aperture radar automatic target recognition (SAR-ATR). It is very important to select discriminative features to train a classifier, which is a prerequisite. Inspired by the great success of convolutional neural network (CNN), we address the problem of SAR target classification by proposing a feature extraction method, which takes advantage of exploiting the extracted deep features from CNNs on SAR images to introduce more powerful discriminative features and robust representation ability for them. First, the pretrained VGG-S net is fine-tuned on moving and stationary target acquisition and recognition (MSTAR) public release database. Second, after a simple preprocessing is performed, the fine-tuned network is used as a fixed feature extractor to extract deep features from the processed SAR images. Third, the extracted deep features are fused by using a traditional concatenation and a discriminant correlation analysis algorithm. Finally, for target classification, K-nearest neighbors algorithm based on LogDet divergence-based metric learning triplet constraints is adopted as a baseline classifier. Experiments on MSTAR are conducted, and the classification accuracy results demonstrate that the proposed method outperforms the state-of-the-art methods.

High-resolution remote sensing data classification has been a challenging and promising research topic in the community of remote sensing. In recent years, with the rapid advances of deep learning, remarkable progress has been made in this field, which facilitates a transition from hand-crafted features designing to an automatic end-to-end learning. A deep fully convolutional networks (FCNs) based ensemble learning method is proposed to label the high-resolution aerial images. To fully tap the potentials of FCNs, both the Visual Geometry Group network and a deeper residual network, ResNet, are employed. Furthermore, to enlarge training samples with diversity and gain better generalization, in addition to the commonly used data augmentation methods (e.g., rotation, multiscale, and aspect ratio) in the literature, aerial images from other datasets are also collected for cross-scene learning. Finally, we combine these learned models to form an effective FCN ensemble and refine the results using a fully connected conditional random field graph model. Experiments on the ISPRS 2-D Semantic Labeling Contest dataset show that our proposed end-to-end classification method achieves an overall accuracy of 90.7%, a state-of-the-art in the field.

Feature extraction techniques are extensively being used in satellite imagery and getting impressive attention for remote sensing applications. The state-of-the-art feature extraction methods are appropriate according to the categories and structures of the objects to be detected. Based on distinctive computations of each feature extraction method, different types of images are selected to evaluate the performance of the methods, such as binary robust invariant scalable keypoints (BRISK), scale-invariant feature transform, speeded-up robust features (SURF), features from accelerated segment test (FAST), histogram of oriented gradients, and local binary patterns. Total computational time is calculated to evaluate the speed of each feature extraction method. The extracted features are counted under shadow regions and preprocessed shadow regions to compare the functioning of each method. We have studied the combination of SURF with FAST and BRISK individually and found very promising results with an increased number of features and less computational time. Finally, feature matching is conferred for all methods.

The tremendous success of deep learning models such as convolutional neural networks (CNNs) in computer vision provides a method for similar problems in the field of remote sensing. Although research on repurposing pretrained CNN to remote sensing tasks is emerging, the scarcity of labeled samples and the complexity of remote sensing imagery still pose challenges. We developed a knowledge-guided golf course detection approach using a CNN fine-tuned on temporally augmented data. The proposed approach is a combination of knowledge-driven region proposal, data-driven detection based on CNN, and knowledge-driven postprocessing. To confront data complexity, knowledge-derived cooccurrence, composition, and area-based rules are applied sequentially to propose candidate golf regions. To confront sample scarcity, we employed data augmentation in the temporal domain, which extracts samples from multitemporal images. The augmented samples were then used to fine-tune a pretrained CNN for golf detection. Finally, commission error was further suppressed by postprocessing. Experiments conducted on GF-1 imagery prove the effectiveness of the proposed approach.

Automatic extraction of buildings from remote sensing data is an attractive research topic, useful for several applications, such as cadastre and urban planning. This is mainly due to the inherent artifacts of the used data and the differences in viewpoint, surrounding environment, and complex shape and size of the buildings. This paper introduces an efficient deep learning framework based on convolutional neural networks (CNNs) toward building extraction from orthoimages. In contrast to conventional deep approaches in which the raw image data are fed as input to the deep neural network, in this paper the height information is exploited as an additional feature being derived from the application of a dense image matching algorithm. As test sites, several complex urban regions of various types of buildings, pixel resolutions and types of data are used, located in Vaihingen in Germany and in Perissa in Greece. Our method is evaluated using the rates of completeness, correctness, and quality and compared with conventional and other “shallow” learning paradigms such as support vector machines. Experimental results indicate that a combination of raw image data with height information, feeding as input to a deep CNN model, provides potentials in building detection in terms of robustness, flexibility, and efficiency.

Recently, unmanned aerial vehicles (UAVs) have gained much attention. In particular, there is a growing interest in utilizing UAVs for agricultural applications such as crop monitoring and management. We propose a computerized system that is capable of detecting Fusarium wilt of radish with high accuracy. The system adopts computer vision and machine learning techniques, including deep learning, to process the images captured by UAVs at low altitudes and to identify the infected radish. The whole radish field is first segmented into three distinctive regions (radish, bare ground, and mulching film) via a softmax classifier and K-means clustering. Then, the identified radish regions are further classified into healthy radish and Fusarium wilt of radish using a deep convolutional neural network (CNN). In identifying radish, bare ground, and mulching film from a radish field, we achieved an accuracy of ≥97.4%. In detecting Fusarium wilt of radish, the CNN obtained an accuracy of 93.3%. It also outperformed the standard machine learning algorithm, obtaining 82.9% accuracy. Therefore, UAVs equipped with computational techniques are promising tools for improving the quality and efficiency of agriculture today.

Semantic segmentation has recently made rapid progress in the field of remote sensing and computer vision. However, many leading approaches cannot simultaneously translate label maps to possible source images with a limited number of training images. The core issue is insufficient adversarial information to interpret the inverse process and proper objective loss function to overcome the vanishing gradient problem. We propose the use of conditional least squares generative adversarial networks (CLS-GAN) to delineate visual objects and solve these problems. We trained the CLS-GAN network for semantic segmentation to discriminate dense prediction information either from training images or generative networks. We show that the optimal objective function of CLS-GAN is a special class of f-divergence and yields a generator that lies on the decision boundary of discriminator that reduces possible vanished gradient. We also demonstrate the effectiveness of the proposed architecture at translating images from label maps in the learning process. Experiments on a limited number of high resolution images, including close-range and remote sensing datasets, indicate that the proposed method leads to the improved semantic segmentation accuracy and can simultaneously generate high quality images from label maps.

High-resolution commercial synthetic aperture radar (SAR) satellites with resolutions of several meters have recently been used for effective disaster monitoring. One study reported the earthquake’s displacement using the pixel matching method with both pre- and postevent TerraSAR-X data, with a validated accuracy of ∼30  cm at global navigation satellite system (GNSS) Earth observation network (GEONET) reference points. However, it is insufficient to determine the accuracy using analysis of only a couple of data points per orbit. In addition, the errors were not reported because the number of data samples was too small to discuss the statistics. In order to better understand displacement accuracy, we analyzed displacement features using the pixel matching method to evaluate the relative geolocation accuracies of the TerraSAR-X product. First, we used fast Fourier transform oversampling 16 times to develop the pixel matching method for estimating the displacement at the subpixel level using the TerraSAR-X StripMap dataset. Second, we applied this methodology to 20 pairs of images from the Tokyo metropolitan area and calculated the displacement for each image pair. Third, we conducted spatial and temporal analyses in order to understand the displacement features. Finally, we evaluated the displacement accuracy by comparison with GEONET and solid earth tide data as a reference.

The visible infrared imaging radiometer suite (VIIRS) is a key sensor carried on the Suomi national polar-orbiting partnership (S-NPP) satellite, which was launched in October 2011. It has several on-board calibration components, including a solar diffuser and a solar diffuser stability monitor for the reflective solar bands, a V-groove blackbody for the thermal emissive bands (TEB), and a space view port for background subtraction. These on-board calibrators are located at fixed scan angles. The VIIRS response versus scan angle (RVS) was characterized prelaunch in lab ambient conditions and is currently used to characterize the on-orbit response for all scan angles relative to the calibrator scan angle. Since the RVS is vitally important to the quality of calibrated radiance products, several independent studies were performed to analyze the prelaunch RVS measurement data. A spacecraft level pitch maneuver was scheduled during the first 3 months of intensive Cal/Val. The S-NPP pitch maneuver provided a rare opportunity for VIIRS to make observations of deep space over the entire range of Earth view scan angles, which can be used to characterize the TEB RVS. This study provides our analysis of the pitch maneuver data and assessment of the derived TEB RVS by comparison with prelaunch results. In addition, the stability of the RVS after the first 5 years of operation is examined using observed brightness temperatures (BT) over a clear ocean at various angles of incidence (AOI). To reduce the impact of variations in the BT measurements, the daily overpasses collected over the ocean are screened for cloud contamination, normalized to the results obtained at the blackbody AOI, and averaged each year.

The performance of a low-cost, self-contained, compact, and easy to deploy mapping-grade mobile laser scanning (MLS) system, which is composed of a light detection and ranging sensor Velodyne VLP-16 and a dual antenna global navigation satellite system/inertial navigation system SBG Systems Ellipse-D, is analyzed. The field tests were carried out in car-mounted and backpack modes for surveying road engineering structures (such as roads, parking lots, underpasses, and tunnels) and coastal erosion zones, respectively. The impact of applied calculation principles on trajectory postprocessing, direct georeferencing, and the theoretical accuracy of the system is analyzed. A calibration method, based on Bound Optimization BY Quadratic Approximation, for finding the boresight angles of an MLS system is proposed. The resulting MLS point clouds are compared with high-accuracy static terrestrial laser scanning data and survey-grade MLS data from a commercially manufactured MLS system. The vertical, horizontal, and relative accuracy are assessed—the root-mean-square error (RMSE) values were determined to be 8, 15, and 3 cm, respectively. Thus, the achieved mapping-grade accuracy demonstrates that this relatively compact and inexpensive self-assembled MLS can be successfully used for surveying the geometry and deformations of terrain, buildings, road, and other engineering structures.

Radiometric calibration of the Landsat-8 operational land imager (OLI) sensor is a key step that enables additional quantitative research and application of the data. This study introduces a ground reflected radiance-based approach to vicarious radiometric validate the Landsat-8 OLI onboard radiometric calibration results. Vicarious radiometric calibration/validation field campaign of the Landsat-8 OLI sensor was carried out at the Baotou site in Inner Mongolia, China, on March 27, 2015, October 28, 2015, March 29, 2016, and April 21, 2016. The relative errors of the OLI bands are within 5% for the ground reflected radiance-based approach when compared with satellite observations of top of atmosphere (TOA) radiance in the OLI bands, and the average 1σ standard deviation of the four results using the vicarious radiometric calibration is less than 0.025. A well-calibrated Terra MODIS sensor was used to cross-validate the Landsat-8 OLI sensor, and the relative errors are within 10%. The results suggest that the OLI onboard the Landsat-8 satellite displays stable radiometric performance for the four calibration/validation days. An uncertainty analysis of the ground reflected radiance-based approach showed an overall uncertainty of less than 4.5%, taking into account uncertainty sources, including surface characteristics, atmospheric characteristics, radiative transfer, solar irradiance, and the calibration model.

Research on the scattering behavior of terahertz (THz) waves plays a prominent part in the development of remote sensing and imaging systems. Due to the relatively short wavelengths of THz waves, materials that could be considered smooth at microwave band may begin to display rough surface in the THz region. To study the scattering characteristics and implement parameter estimation and imaging of rough surface rotating targets in the THz band, THz radar systems with carrier frequencies of 220 and 440 GHz were built, and experiments based on a series of rough surface cylinders were carried out. From the time–frequency distributions and imaging results, the transition of scattering characteristics from smooth surface to rough surface is clearly visible, which is consistent with the fact that as the roughness increases, there is an overall increase in diffuse scattering. In addition, methods of rotation period estimation and target size estimation were proposed and applied to experimental data of the THz radar systems, and their validities were verified by the high-precision estimation results.

Considering the sparseness of scatterers in the scene of a synthetic aperture radar (SAR) image, we propose a modified model for SAR images with enhanced features by automatically choosing variable lk-norm and regularization parameter. The approach is based on a regularized reconstruction of the scattering field, which employs prior information of the region of interest. It leads to an alternating iterative algorithm for the modeling. The method is constructed based on variable lk-norm and regularization parameter. Here, k is a function of the imaged region and it could be estimated during the iteration process to the scattering field. The regularization parameter is changing because it is being determined by k. Moreover, the parameter estimators of the presented model are derived by applying the method of log cumulants-based on Mellin transform. Compared to conventional SAR regularization methods, the proposed method reconstructs images with increased resolution, reduced clutter, and reduced computation cost. We demonstrate the performance of the method on real SAR scenes. The experiment results of measured SAR data prove the effectiveness.

Soil moisture plays an important role in the water cycle within the surface ecosystem, and it is the basic condition for the growth of plants. Currently, the spatial resolutions of most soil moisture data from remote sensing range from ten to several tens of km, while those observed in-situ and simulated for watershed hydrology, ecology, agriculture, weather, and drought research are generally <1  km. Therefore, the existing coarse-resolution remotely sensed soil moisture data need to be downscaled. This paper proposes a universal and multitemporal soil moisture downscaling method suitable for large areas. The datasets comprise land surface, brightness temperature, precipitation, and soil and topographic parameters from high-resolution data and active/passive microwave remotely sensed essential climate variable soil moisture (ECV_SM) data with a spatial resolution of 25 km. Using this method, a total of 288 soil moisture maps of 1-km resolution from the first 10-day period of January 2003 to the last 10-day period of December 2010 were derived. The in-situ observations were used to validate the downscaled ECV_SM. In general, the downscaled soil moisture values for different land cover and land use types are consistent with the in-situ observations. Mean square root error is reduced from 0.070 to 0.061 using 1970 in-situ time series observation data from 28 sites distributed over different land uses and land cover types. The performance was also assessed using the GDOWN metric, a measure of the overall performance of the downscaling methods based on the same dataset. It was positive in 71.429% of cases, indicating that the suggested method in the paper generally improves the representation of soil moisture at 1-km resolution.

The detailed three-dimensional modeling of buildings utilizing elevation data, such as those provided by light detection and ranging (LiDAR) airborne scanners, is increasingly demanded today. There are certain application requirements and available datasets to which any research effort has to be adapted. Our dataset includes aerial orthophotos, with a spatial resolution 20 cm, and a digital surface model generated from LiDAR, with a spatial resolution 1 m and an elevation resolution 20 cm, from an area of Athens, Greece. The aerial images are fused with LiDAR, and we classify these data with a multilayer feedforward neural network for building block extraction. The innovation of our approach lies in the preprocessing step in which the original LiDAR data are super-resolution (SR) reconstructed by means of a stochastic regularized technique before their fusion with the aerial images takes place. The Lorentzian estimator combined with the bilateral total variation regularization performs the SR reconstruction. We evaluate the performance of our approach against that of fusing unprocessed LiDAR data with aerial images. We present the classified images and the statistical measures confusion matrix, kappa coefficient, and overall accuracy. The results demonstrate that our approach predominates over that of fusing unprocessed LiDAR data with aerial images.

Identification of water extents in Landsat images is challenging due to surfaces with similar reflectance to water extents. The objective of this study is to provide stable and accurate methods for identifying water extents in Landsat images based on meta-heuristic algorithms. Then, seven Landsat images were selected from various environmental regions in Iran. Training of the algorithms was performed using 40 water pixels and 40 nonwater pixels in operational land imager images of Chitgar Lake (one of the study regions). Moreover, high-resolution images from Google Earth were digitized to evaluate the results. Two approaches were considered: index-based and artificial intelligence (AI) algorithms. In the first approach, nine common water spectral indices were investigated. AI algorithms were utilized to acquire coefficients of optimal band combinations to extract water extents. Among the AI algorithms, the artificial neural network algorithm and also the ant colony optimization, genetic algorithm, and particle swarm optimization (PSO) meta-heuristic algorithms were implemented. Index-based methods represented different performances in various regions. Among AI methods, PSO had the best performance with average overall accuracy and kappa coefficient of 93% and 98%, respectively. The results indicated the applicability of acquired band combinations to extract accurately and stably water extents in Landsat imagery.

Hyperspectral image classification faces a serious challenge due to the high dimension of hyperspectral data and limited labeled samples. Tri-training algorithm is a widely used semisupervised classification method, but the algorithm lacks significant diversity among the classifiers when the number of initial label samples is limited. A semisupervised classification method for hyperspectral data based on tri-training is proposed. It combines different classifiers and stratified sampling based on labeled class to increase classifier diversity and avoid classifier performance deterioration. Performance comparison between the proposed algorithm and tri-training algorithm was made through experiments. The proposed algorithm improved the overall accuracy and Kappa coefficient by 1.37% to 6.84% and 0.0096 to 0.0808, respectively, and the results showed that the effectiveness of the algorithm is verified. Moreover, the algorithm can also get better performance when the number of samples is small.

By taking advantages of fine details obtained by the improved spatial resolution, very high-resolution images are promising for detecting change regions and identifying change patterns. However, high overlaps between different change patterns and the complexities of multiclass classification make it difficult to reliably separate change features. A framework named simultaneous change region and pattern identification (SCRAPI) is proposed for simultaneously detecting change regions and identifying change patterns, whose components are aimed at capturing overlaps between change patterns and reducing overlaps driven by user-specific interests. To validate the effectiveness of SCRAPI, a supervised approach is illustrated within this framework, which starts with modeling the relationship between change features by interclass couples and intraclass couples, followed by metric learning where structural sparsity is captured by the mixed norm. Experiments demonstrate the effectiveness of the proposed approach.

Recent advances in camouflage engineering have made it more difficult to detect targets. Assessing the effectiveness of camouflage against different target detection methods leads to identifying the strengths and weaknesses of camouflage designs. One of the target detection methods is to analyze the content of the scene using remote sensing hyperspectral images. In the process of evaluating camouflage designs, there must be comprehensive and efficient evaluation criteria. Three parameters were considered as the main factors affecting the target detection and based on these factors, camouflage effectiveness assessment criteria were proposed. To combine the criteria in the form of a single equation, the equation used in target visual search models was employed and for determining the criteria, a model was presented based on the structure of the computational visual attention systems. Also, in software implementations on the HyMap hyperspectral image, a variety of camouflage levels were created for the real targets in the image. Assessing the camouflage levels using the proposed criteria, comparing and analyzing the results can show that the provided criteria and model are effective for the evaluation of camouflage designs using hyperspectral images.

An ensemble framework for multiple-instance (MI) learning (MIL) is introduced for use in hyperspectral images (HSIs) by inspiring the bagging (bootstrap aggregation) method in ensemble learning. Ensemble-based bagging is performed by a small percentage of training samples, and MI bags are formed by a local windowing process with variable window sizes on selected instances. In addition to bootstrap aggregation, random subspace is another method used to diversify base classifiers. The proposed method is implemented using four MIL classification algorithms. The classifier model learning phase is carried out with MI bags, and the estimation phase is performed over single-test instances. In the experimental part of the study, two different HSIs that have ground-truth information are used, and comparative results are demonstrated with state-of-the-art classification methods. In general, the MI ensemble approach produces more compact results in terms of both diversity and error compared to equipollent non-MIL algorithms.

A multifeature soft-probability cascading scheme to solve the problem of land use and land cover (LULC) classification using high-spatial-resolution images to map rural residential areas in China is proposed. The proposed method is used to build midlevel LULC features. Local features are frequently considered as low-level feature descriptors in a midlevel feature learning method. However, spectral and textural features, which are very effective low-level features, are neglected. The acquisition of the dictionary of sparse coding is unsupervised, and this phenomenon reduces the discriminative power of the midlevel feature. Thus, we propose to learn supervised features based on sparse coding, a support vector machine (SVM) classifier, and a conditional random field (CRF) model to utilize the different effective low-level features and improve the discriminability of midlevel feature descriptors. First, three kinds of typical low-level features, namely, dense scale-invariant feature transform, gray-level co-occurrence matrix, and spectral features, are extracted separately. Second, combined with sparse coding and the SVM classifier, the probabilities of the different LULC classes are inferred to build supervised feature descriptors. Finally, the CRF model, which consists of two parts: unary potential and pairwise potential, is employed to construct an LULC classification map. Experimental results show that the proposed classification scheme can achieve impressive performance when the total accuracy reached about 87%.

The narrowband interference (NBI) can degrade the synthetic aperture radar (SAR) imaging quality severely. This paper proposes an NBI mitigation method using the variational mode decomposition (VMD). The coarse estimation of NBI is obtained by decomposing the real part and imaginary part of the complex-valued raw echoes into a number of modes by VMD independently. Next, modes that correspond to NBI are refined by the mask technique in the frequency domain. Then the interference is mitigated by subtracting the refined estimated NBI components from the echoes, and a well-focused SAR image is obtained by conventional imaging schemes. The proposed method outperforms other time-varying NBI mitigation methods with smaller effective data loss and less impact on the focusing performance of images. Results of simulated and measured data prove the validity of the proposed method.

Multisensor image fusion can be used as an advanced technique for image enhancement. A fusion method based on region division strategy is proposed for synthetic aperture radar (SAR) and color visible images. The SAR target feature detection function is first constructed to divide the images into the SAR target region and non-SAR target region. Meanwhile, SAR image despeckling is considered during the generation of this detection function. Then, two different fusion rules are designed for these two regions. More specifically, the fusion rule adopted in the SAR target region is fusing the high-frequency feature within the SAR image, and the fusion rule employed in the non-SAR target region is maintaining the spectral and detail information of the color visible image. Experimental results on several pairs of SAR and color visible images demonstrate the effectiveness of the proposed method. Compared with the conventional image fusion methods, the proposed method provides better results for all evaluation criteria, including spectral distortions, feature similarity, mutual information, and peak signal-to-noise ratio.

The performance of space–time adaptive processing (STAP) may degrade significantly when some of the training samples are contaminated by the signal-like components (outliers) in nonhomogeneous clutter environments. To remove the training samples contaminated by outliers in nonhomogeneous clutter environments, a robust nonhomogeneous training samples detection method using the sparse-recovery (SR) with knowledge-aided (KA) is proposed. First, the reduced-dimension (RD) overcomplete spatial–temporal steering dictionary is designed with the prior knowledge of system parameters and the possible target region. Then, the clutter covariance matrix (CCM) of cell under test is efficiently estimated using a modified focal underdetermined system solver (FOCUSS) algorithm, where a RD overcomplete spatial–temporal steering dictionary is applied. Third, the proposed statistics are formed by combining the estimated CCM with the generalized inner products (GIP) method, and the contaminated training samples can be detected and removed. Finally, several simulation results validate the effectiveness of the proposed KA-SR-GIP method.

We present a technique for interferometric processing of extreme high-squint synthetic aperture radar (SAR) target images. The technique is an improvement over conventional SAR interferometry, as it takes into account the change in relative phase between channels that occur as the radar moves through the synthetic aperture, which can be significant for forward-looking data acquisition geometry. We apply the technique to simulate data from a variety of imaging cases and show that, unlike the conventional interferometric technique, the proposed technique is robust to synthetic aperture length, scatterer elevation, and scatterer dimension in forward-looking scenarios. The technique may be used for various applications related to situational awareness, such as remote sensing for collision avoidance or forward-looking reconnaissance.

To accelerate the massive remote sensing images (RSIs) coding in a ground service-oriented remote sensing system, this study proposes three-level (i.e., tree-level, bit-plane level, and byte-level) parallel-set partitioning in hierarchical trees (TP-SPIHT) coding on a collaborative central and graphic processing unit (CPU and GPU) to parallelize SPIHT by optimizing its dynamic processing with the linked list. Basic parallel SPIHT coding is presented with preprocessing, tree-level parallel coding, and bit-stream organization using three kinds of static marker matrices instead of the dynamic linked lists originally used to remove the data dependency of the original SPIHT. The bit-stream organization is implemented on CPU and other processes are implemented on GPU using GPU streams. The bit-stream organization is further divided into a bit-plane level parallel bit-plane stream extraction and a final bit-stream organization on a multicore CPU. Because no dependencies exist between the different byte operations in the final bit-stream organization, this organization is accelerated by byte-level parallelization on the GPU. Experimental results with different sized RSIs show that TP-SPIHT takes 292.03 ms to code a 2048×2048 image and achieves a 6.27 times speedup compared with an optimized CPU implementation. The speedup ratio improves as the image increases from 256×256 to 2048×2048.

The key question of sparse coding (SC) is how to exploit the information that already exists to acquire the robust sparse representations (SRs) of distinguishing different objects for hyperspectral image (HSI) classification. We propose a multi-information fusion SC framework, which fuses the spectral, spatial, and label information in the same level, to solve the above question. In particular, pixels from disjointed spatial clusters, which are obtained by cutting the given HSI in space, are individually and sparsely encoded. Then, due to the importance of spatial structure, graph- and hypergraph-based regularizers are enforced to motivate the obtained representations smoothness and to preserve the local consistency for each spatial cluster. The latter simultaneously considers the spectrum, spatial, and label information of multiple pixels that have a great probability with the same label. Finally, a linear support vector machine is selected as the final classifier with the learned SRs as input. Experiments conducted on three frequently used real HSIs show that our methods can achieve satisfactory results compared with other state-of-the-art methods.

The Doppler lidar technology is known for its ability to measure accurate winds with fine time and space resolutions. The derivation of turbulence parameters from lidar wind measurement has been attempted by several authors. All of them relate the turbulence parameters to long-time series (several tens of minutes) of wind measurements. The method presented here retrieves estimations of the atmospheric turbulence at much finer time scales. The technique is based on a wind reconstruction method applied to a five-beam wind Doppler lidar (namely the WindCube model by Leosphere). The method relies on a particle filter. The suggested reconstruction algorithm links the lidar observations to numerical particles to obtain turbulence estimations every time new observations are available. The high frequency of the estimations is an innovation and is detailed and discussed here. Moreover, the presented algorithm enables reconstruction of the wind in three dimensions in the observed volume. Thus, we locally have access to the spatial variability of the turbulent atmosphere. The suggested algorithm is applied to a set of real observations. The results show that the estimation of the turbulent parameters is significantly improved. They open the way to the use of lidars for scientific and industrial purposes such as site studies for wind farms.

Water monitoring is an important part of water resource management and has become an essential aspect of remote sensing. A number of indices have been developed for water extraction using satellite images. Even though all indices can extract the extent of a water body, none can do so without including a noise component, such as topographic shadows, cloud shadows, snow, ice, and buildup areas, all of which have spectrally similar characteristics under certain circumstances. In order to select the best index for water body extraction, several water indices have been compared. This paper proposes a method for extracting water bodies called the water extraction surface temperature index (WESTI). This method uses normalized difference water index (NDWI) and land surface temperature to eliminate the noise components, especially in mountainous and cold areas where other indices have very low accuracy. The results have shown that WESTI improves the NDWI results by removing more than 80% of topographic shadows, with an overall accuracy of 99% in all cases.

Updated and accurate information of rice-growing areas is vital for food security and investigating the environmental impact of rice ecosystems. The intent of this work is to explore the feasibility of dual-polarimetric C-band Radar Imaging Satellite-1 (RISAT-1) data in delineating rice crop fields from other land cover features. A two polarization combination of RISAT-1 backscatter, namely ratio (HH/HV) and difference (HH−HV), significantly enhanced the backscatter difference between rice and nonrice categories. With these inputs, a QUEST decision tree (DT) classifier is successfully employed to extract the spatial distribution of rice crop areas. The results showed the optimal polarization combination to be HH along with HH/HV and HH−HV for rice crop mapping with an accuracy of 88.57%. Results were further compared with a Landsat-8 operational land imager (OLI) optical sensor-derived rice crop map. Spatial agreement of almost 90% was achieved between outputs produced from Landsat-8 OLI and RISAT-1 data. The simplicity of the approach used in this work may serve as an effective tool for rice crop mapping.

Reducing the spectral dimension of hyperspectral data without loss of information is an important process in hyperspectral imagery. By adding the collaborative reconstructive information in linear discriminant analysis (LDA), we propose a discriminant collaborative neighborhood preserving embedding (DCNPE) for feature extraction from hyperspectral images. In the proposed DCNPE, an l2-graph is constructed based on the collaborative representation (CR). The edge weights of graph are calculated by l2-norm minimization using all samples to represent the pointed sample. The proposed DCNPE method aims to find a projection, which can preserve the CR-based reconstruction relationship of data as well as maximize the class discrimination of the data. DCNPE inherits the advantages of both LDA and CR. It not only overcomes the reduced dimension limitation of LDA but also preserves the collaborative neighborhood relations of data through l2-graph. Experimental results on three real HSI datasets certify its effectiveness of dimensionality reduction by showing a better classification performance.

Coffee is an important crop in tropical regions of the world; about 125 million people depend on coffee agriculture for their livelihoods. Understanding the spatial extent of coffee fields is useful for management and control of coffee pests such as Hypothenemus hampei and other pests that use coffee fruit as a host for immature stages such as the Mediterranean fruit fly, for economic planning, and for following changes in coffee agroecosystems over time. We present two methods for detecting Coffea arabica fields using remote sensing and geospatial technologies on WorldView-2 high-resolution spectral data of the Kona region of Hawaii Island. The first method, a pixel-based method using a maximum likelihood algorithm, attained 72% producer accuracy and 69% user accuracy (68% overall accuracy) based on analysis of 104 ground truth testing polygons. The second method, an object-based image analysis (OBIA) method, considered both spectral and textural information and improved accuracy, resulting in 76% producer accuracy and 94% user accuracy (81% overall accuracy) for the same testing areas. We conclude that the OBIA method is useful for detecting coffee fields grown in the open and use it to estimate the distribution of about 1050 hectares under coffee agriculture in the Kona region in 2012.

Advanced spaceborn thermal emission and reflection radiometer (ASTER) has fine spectral bands in short-wave infrared (SWIR) and thermal infrared (TIR) regions of the electromagnetic spectrum. The purpose behind the study is to explore the potential of ASTER for lithological and minerals detection; in comparison with Landsat-ETM+ Khaira Murat range (KMR) of Gali Jagir area, district Attock was selected as a test site; enriched with industrial minerals of the Eocene age. Maximum likelihood classification was applied on Landsat-ETM+ and ASTER images. Maximum likelihood classification on ASTER satellite image exhibits better discrimination among various lithologies as compared to Landsat-ETM+. Classified image of ASTER showed a correlation coefficient of 0.6 with the geological survey of Pakistan’s map while a classified image of Landsat-ETM+ exhibited a correlation of only 0.43. Landsat-ETM+ and ASTER satellite images were further investigated for minerals detection. Landsat-ETM+ band ratio detected clay. ASTER SWIR band ratios detected various clay and carbonate minerals. X’PertPRO diffractometer and differential thermal analysis of field samples verified the detected minerology. The results suggest that ASTER can be successfully used for lithological and minerals mapping of less-examined areas.

Convolutional neural network (CNN), as a vital part of the deep learning research field, has shown powerful potential for automatic target recognition (ATR) of synthetic aperture radar (SAR). However, the high complexity caused by the deep structure of CNN makes it difficult to generalize. An improved form of CNN with higher generalization capability and less probability of overfitting, which further improves the efficiency and robustness of the SAR ATR system, is proposed. The convolution layers of CNN are combined with a two-dimensional principal component analysis algorithm. Correspondingly, the kernel support vector machine is utilized as the classifier layer instead of the multilayer perceptron. The verification experiments are implemented using the moving and stationary target acquisition and recognition database, and the results validate the efficiency of the proposed method.

Traditional surface water quality assessment is costly, labor intensive, and time consuming; however, remote sensing has the potential to assess surface water quality because of its spatiotemporal consistency. Therefore, estimating concentrations of surface water quality parameters (SWQPs) from satellite imagery is essential. Remote sensing estimation of nonoptical SWQPs, such as chemical oxygen demand (COD), biochemical oxygen demand (BOD), and dissolved oxygen (DO), has not yet been performed because they are less likely to affect signals measured by satellite sensors. However, concentrations of nonoptical variables may be correlated with optical variables, such as turbidity and total suspended sediments, which do affect the reflected radiation. In this context, an indirect relationship between satellite multispectral data and COD, BOD, and DO can be assumed. Therefore, this research attempts to develop an integrated Landsat 8 band ratios and stepwise regression to estimate concentrations of both optical and nonoptical SWQPs. Compared with previous studies, a significant correlation between Landsat 8 surface reflectance and concentrations of SWQPs was achieved and the obtained coefficient of determination (R2)&gt;0.85. These findings demonstrated the possibility of using our technique to develop models to estimate concentrations of SWQPs and to generate spatiotemporal maps of SWQPs from Landsat 8 imagery.

The Colorado front range (CFR) mountains have a history of significant debris flows. For instance, flash floods in September 2013 triggered over 1138 debris flows in the CFR, leading to eight fatalities and causing damage to buildings, highways, railroads, and infrastructure. Following this event, the United States Geological Survey (USGS) studied these debris flows with the intention of modeling debris flow susceptibility in this region. They identified a problem that the preliminary susceptibility map, before correction, included areas of exposed bedrock as regions that were modeled to have high susceptibility, even though those regions should have been low susceptibility. The objective of this research is to develop and execute an integrated sequential land cover classification (iSLCC) remote sensing approach to discriminate bedrock outcrop from colluvium and, hence, constrain the USGS susceptibility modeling. The approach integrates multispectral, hyperspectral, and radar methods to discriminate colluvium from bedrock outcrops over a study area of nine 7½ min quadrangles. Calibrating against six smaller field study areas that span different geologic formations and ecoregions, a map of land cover, including exposed bedrock outcrops, was produced over a study area that encompasses portions of the St. Vrain and Big Thompson watersheds. The resulting map was compared with the results from traditional land cover classification methods. The iSLCC approach yielded the best overall results and the highest observed agreement (observed accuracy).

Hyperspectral images may be applied to classify objects in a scene. The redundancy in hyperspectral data implies that fewer spectral features might be sufficient for discriminating the objects captured in a scene. The availability of labeled classes of several areas in a scene paves the way for a supervised dimensionality reduction, i.e., using a discrimination measure between the classes in a scene to select spectral features. We show that averaging adjacent spectral channels and using wider spectral regions yield a better class separability than the selection of individual channels from the original hyperspectral dataset. We used a method named spectral region splitting (SRS), which creates a new feature space by averaging neighboring channels. In addition to the common benefits of channel selection methods, the algorithm constructs wider spectral regions when it is useful. Using different class separability measures over various datasets resulted in a better discrimination between the classes than the best-selected channels using the same measure. The reason is that the wider spectral regions led to a reduction in intraclass distances and an improvement in class discrimination. The overall classification accuracy of two hyperspectral scenes gave an increase of about two-percent when using the spectral regions determined by applying SRS.

The forward-scattering model is introduced to describe the response of normalized radar cross section (NRCS) of precipitation with synthetic aperture radar (SAR). Since the distribution of near-surface rainfall is related to the rate of near-surface rainfall and horizontal distribution factor, a retrieval algorithm called modified regression empirical and model-oriented statistical (M–M) based on the volterra integration theory is proposed. Compared with the model-oriented statistical and volterra integration (MOSVI) algorithm, the biggest difference is that the M–M algorithm is based on the modified regression empirical algorithm rather than the linear regression formula to retrieve the value of near-surface rainfall rate. Half of the empirical parameters are reduced in the weighted integral work and a smaller average relative error is received while the rainfall rate is less than 100  mm/h. Therefore, the algorithm proposed in this paper can obtain high-precision rainfall information.

Evapotranspiration (ET) plays an important role in surface–atmosphere interactions and can be monitored using remote sensing data. The visible infrared imaging radiometer suite (VIIRS) sensor is a generation of optical satellite sensors that provide daily global coverage at 375- to 750-m spatial resolutions with 22 spectral channels (0.412 to 12.05  μm) and capable of monitoring ET from regional to global scales. However, few studies have focused on methods of acquiring ET from VIIRS images. The objective of this study is to introduce an algorithm that uses the VIIRS data and meteorological variables to estimate the energy budgets of land surfaces, including the net radiation, soil heat flux, sensible heat flux, and latent heat fluxes. A single-source model that based on surface energy balance equation is used to obtain surface heat fluxes within the Zhangye oasis in China. The results were validated using observations collected during the HiWATER (Heihe Watershed Allied Telemetry Experimental Research) project. To facilitate comparison, we also use moderate resolution imaging spectrometer (MODIS) data to retrieve the regional surface heat fluxes. The validation results show that it is feasible to estimate the turbulent heat flux based on the VIIRS sensor and that these data have certain advantages (i.e., the mean bias error of sensible heat flux is 15.23  W m−2) compared with MODIS data (i.e., the mean bias error of sensible heat flux is −29.36  W m−2). Error analysis indicates that, in our model, the accuracies of the estimated sensible heat fluxes rely on the errors in the retrieved surface temperatures and the canopy heights.

A precise estimate of canopy leaf nitrogen concentration (CNC, based on dry mass) is important for researching the carbon assimilation capability of forest ecosystems. Hyperspectral remote sensing technology has been applied to estimate regional CNC, which can adjust forest photosynthetic capacity and carbon uptake. However, the relationship between forest CNC and canopy spectral reflectance as well as its mechanism is still poorly understood. Using measured CNC, canopy structure and species composition data, four vegetation indices (VIs), and near-infrared reflectance (NIR) derived from EO-1 Hyperion imagery, we investigated the role of canopy structure traits and plant functional types (PFTs) in modulating the correlation between CNC and canopy reflectance in a temperate forest in northeast China. A plot-scale forest structure indicator, named broad foliar dominance index (BFDI), was introduced to provide forest canopy structure and coniferous and broadleaf species composition. Then, we revealed the response of forest canopy reflectance spectrum to BFDI and CNC. Our results showed that leaf area index had no significant effect on NIR ( P&gt;0.05) but indicated that there was a significant correlation ( R2=0.76, P&lt;0.0001) between CNC and BFDI. NIR had a more significant correlation with BFDI than with CNC for all PFTs, but it had no obvious correlation with CNC for single PFT. Partial correlation analysis showed that four VIs had better correlations with BFDI than with CNC. When the effect of BFDI was removed, the partial correlation between CNC and NIR was insignificant ( R=0.273, P&gt;0.05). On the contrary, removing the CNC effect, the partial correlation between BFDI and NIR was positively significant ( R=0.69, P&lt;0.0001). These findings proved that canopy structure and coniferous and broadleaf species composition had a greater influence on the remote sensing signal than canopy nitrogen concentration. The functional convergence of plant traits resulted in the relation of CNC and canopy structure and determined the positive correlation between CNC and NIR. We maintain that the repeatable relationship between CNC and NIR can be used in the remote sensing retrieval of CNC during various forest types. Nevertheless, the relationship cannot b

Field-portable goniometers are created for a wide variety of applications. Many of these applications require specific types of instruments and measurement schemes and must operate in challenging environments. Therefore, designs are based on the requirements that are specific to the application. We present a field-portable goniometer that was designed for measuring the hemispherical-conical reflectance factor (HCRF) of various soils and low-growing vegetation in austere coastal and desert environments and biconical reflectance factors in laboratory settings. Unlike some goniometers, this system features a requirement for “target-plane tracking” to ensure that measurements can be collected on sloped surfaces, without compromising angular accuracy. The system also features a second upward-looking spectrometer to measure the spatially dependent incoming illumination, an integrated software package to provide full automation, an automated leveling system to ensure a standard frame of reference, a design that minimizes the obscuration due to self-shading to measure the opposition effect, and the ability to record a digital elevation model of the target region. This fully automated and highly mobile system obtains accurate and precise measurements of HCRF in a wide variety of terrain and in less time than most other systems while not sacrificing consistency or repeatability in laboratory environments.

Land use and land cover (LULC) data are important to monitor and assess environmental change. LULC classification using satellite images is a method widely used on a global and local scale. Especially, urban areas that have various LULC types are important components of the urban landscape and ecosystem. This study aims to classify urban LULC using WorldView-3 (WV-3) very high-spatial resolution satellite imagery and the object-based image analysis method. A decision rules set was applied to classify the WV-3 images in Kathu subdistrict, Phuket province, Thailand. The main steps were as follows: (1) the image was ortho-rectified with ground control points and using the digital elevation model, (2) multiscale image segmentation was applied to divide the image pixel level into image object level, (3) development of the decision ruleset for LULC classification using spectral bands, spectral indices, spatial and contextual information, and (4) accuracy assessment was computed using testing data, which sampled by statistical random sampling. The results show that seven LULC classes (water, vegetation, open space, road, residential, building, and bare soil) were successfully classified with overall classification accuracy of 94.14% and a kappa coefficient of 92.91%.

Land surface temperature (LST) is one of the key states of the Earth surface system. Remote sensing has the capability to obtain high-frequency LST observations with global coverage. However, mainly due to cloud cover, there are always gaps in the remotely sensed LST product, which hampers the application of satellite-based LST in data-driven modeling of surface energy and water exchange processes. We explored the suitability of the data interpolating empirical orthogonal functions (DINEOF) method in moderate resolution imaging spectroradiometer LST reconstruction around Ali on the Tibetan Plateau. To validate the reconstruction accuracy, synthetic clouds during both daytime and nighttime are created. With DINEOF reconstruction, the root mean square error and bias under synthetic clouds in daytime are 4.57 and −0.0472  K, respectively, and during the nighttime are 2.30 and 0.0045 K, respectively. The DINEOF method can well recover the spatial pattern of LST. Time-series analysis of LST before and after DINEOF reconstruction from 2002 to 2016 shows that the annual and interannual variabilities of LST can be well reconstructed by the DINEOF method.

Snow sublimation is an important hydrological process and its spatial and temporal variation remains largely unknown; however, few studies have been conducted to quantify its spatial variability. Our study focuses on the evaluation of two algorithms, Penman–Monteith (P–M) equation and the bulk aerodynamic (BA) parameterization of snow sublimation. The two methods were first evaluated against eddy covariance (EC) measurements of latent heat flux at towers located in the upper reaches of the Heihe River Basin (China). Both methods were in good agreement with the ground observations with high coefficient of determination (<inline-formula> <math display="inline" xmlns:mml="http://www.w3.org/1998/Math/MathML"> <mrow> <msup> <mi>R</mi> <mn>2</mn> </msup> </mrow> </math> </inline-formula>) and low root mean squared error (RMSE). Next, we estimated subpixel snow sublimation using remote sensing data at a <inline-formula> <math display="inline" xmlns:mml="http://www.w3.org/1998/Math/MathML"> <mrow> <mn>1</mn> <mtext>-</mtext> <mi>km</mi> <mo>×</mo> <mn>1</mn> <mtext>-</mtext> <mi>km</mi> </mrow> </math> </inline-formula> spatial resolution. The results based on satellite data were evaluated against ground measurements at the two experimental sites. The P–M equation gave <inline-formula> <math display="inline" xmlns:mml="http://www.w3.org/1998/Math/MathML"> <mrow> <msup> <mi>R</mi> <mn>2</mn> </msup> <mo>=</mo> <mn>0.75</mn> </mrow> </math> </inline-formula>, <inline-formula> <math display="inline" xmlns:mml="http://www.w3.org/1998/Math/MathML"> <mrow> <mi>RMSE</mi> <mo>=</mo> <mn>8.4</mn> <mtext>  </mtext> <mi mathvariant="normal">W</mi> <mtext> </mtext> <msup> <mrow> <mi mathvariant="normal">m</mi> </mrow> <mrow> <mo>−</mo> <mn>2</mn> </mrow> </msup> </mrow> </math> </inline-formula> for Dashalong site and <inline-formula> <math display="inline" xmlns:mml="http://www.w3.org/1998/Math/MathML"> <mrow> <msup> <mi>R</mi> <mn>2</mn> </msup> <mo>=</mo> <mn>0.36</mn> </mrow> </math> </inline-formula>, <inline-formula> <math display="inline" xmlns:mml="http://www.w3.org/1998/Math/MathML"> <mrow> <mi>RMSE</mi> <mo>=</mo> <mn>9.1</mn> <mtext>  </mtext> <mi mathvariant="normal">W</mi> <mtext> </mtext> <msup> <mrow> <mi mathvariant="normal">m</mi> </mrow> <mrow> <mo>−</mo> <mn>2</mn> </mrow> </msup> </mrow> </math> </inline-formula> for the Dadongshu site and performed better than the BA parameterization, which gave <inline-formula> <math display="inline" xmlns:mml="http://www.w3.org/1998/Math/MathML"> <mrow> <msup> <mi>R</mi> <mn>2</mn> </msup> <mo>=</mo> <mn>0.65</mn> </mrow> </math> </inline-formula>, <inline-formula> <math display="inline" xmlns:mml="http://www.w3.org/1998/Math/MathML"> <mrow> <mi>RMSE</mi> <mo>=</mo> <mn>17.5</mn> <mtext>  </mtext> <mi mathvariant="normal">W</mi>

Electron content in the ionosphere is very sensitive to temporary disturbances of the Earth’s magnetosphere (geomagnetic storm), solar flares, and seismic activities. The Global Navigation Satellite System (GNSS)-based total electron content (TEC) measurement has emerged as an important technique for computations of earthquake precursor signals. We examined the pre-earthquake signatures for eight strong magnitude ( Mw&gt;6: 6.1 to 7.8) earthquakes with the aid of GNSS-based TEC measurement in the tectonically active Himalayan region using International GNSS Service (IGS) stations as well as local GNSS-based continuously operating reference stations (CORS). The results indicate very significant ionospheric anomalies in the vertical total electron content (vTEC) a few days before the main shock for all of the events. Geomagnetic activities were also studied during the TEC observation window to ascertain their role in ionospheric perturbations. It was also inferred that TEC variation due to low magnitude events could also be monitored if the epicenter lies closer to the GNSS or IGS station. Therefore, the study has confirmed TEC anomalies before major Himalayan earthquakes, thereby making it imperative to set up a much denser network of IGS/CORS for real-time data analysis and forewarning.

Aircraft roll is an important reason for geometric distortion of airborne push-broom hyperspectral images. A two-step rectification method is proposed to correct the geometric distortions of push-broom hyperspectral images caused by aircraft roll. According to the distortion characteristics of hyperspectral images, image distortion is mainly classified into two categories: high frequency distortion and large amplitude distortion. For high frequency distortion, the first-step correction is implemented by slide matching of segmented windows between adjacent scan lines. For large amplitude distortion, the second-step correction is achieved using ground control points and linear features between a reference map and an uncorrected image. Experimental results show that the proposed method can eliminate image distortion caused by aircraft roll effectively.

The Amazon–Cerrado transition forest is an extensive region with unique characteristics of radiation exchanges. The measurements of the net radiation (Rn) in this ecosystem are limited to the local scale, and their spatial distribution can be carried out by remote sensing techniques, of which accuracy needs to be evaluated. Thus, the objective of this study was to analyze the accuracy of the model of surface Rn derived from measured solar radiation and estimates of normalized difference vegetation index (NDVI), surface albedo ( α), and land surface temperature (LST) estimated by images of Landsat 5 TM in an Amazon–Cerrado transition forest. The Rn, NDVI, α, and LST were estimated by Landsat 5 TM images and related to micrometeorological measurements in a tower of the study area. There was seasonality of micrometeorological variables with higher values of incident solar radiation, air temperature, and vapor pressure deficit during the dry season. However, there was no seasonality of Rn. NDVI decreased and α increased during the dry season, while LST was nearly constant. The Rn had negative correlation with α and positive with NDVI. Both instantaneous and daily Rn estimated with Landsat 5 TM images showed high correlation and low error values when compared with Rn measured in the study area.

It is the intent of this paper to demonstrate the veracity of a method to estimate the total solid contaminant mass present on a sparsely coated surface encompassing a pixel from longwave infrared spectra. Sparsely coated surfaces create complex radiometry due to the interactions of electromagnetic energy between intimate materials. Current algorithms can be used on intimate mixtures to identify the abundances of materials in a pixel, but they cannot provide additional property information. Radiative transfer models can create mixture signatures, but only with a set of well-characterized physical parameters that are typically not known or are difficult to retrieve. The approach described here creates a parameter inversion model from a radiative transfer model and uses empirically measured mixture data to retrieve physical characteristics of the contaminated surface and derive a total contaminant mass present.

A simulator for spotlight synthetic aperture radar (SAR) image formation is presented. The simulator produces radar returns from a virtual radar positioned at an arbitrary distance and altitude. The radar returns are produced from a source image, where the return is a weighted summation of linear frequency-modulated (LFM) pulse signals delayed by the distance of each pixel in the image to the radar. The imagery is resampled into polar format to ensure consistent range profiles to the position of the radar. The SAR simulator provides a capability enabling the objective analysis of formed SAR imagery, comparing it to an original source image. This capability allows for analysis of various SAR signal processing techniques previously determined by impulse response function (IPF) analysis. The results suggest that IPF analysis provides results that may not be directly related to formed SAR image quality. Instead, the SAR simulator uses image quality metrics, such as peak signal-to-noise ratio (PSNR) and structured similarity index (SSIM), for formed SAR image quality analysis. To showcase the capability of the SAR simulator, it is used to investigate the performance of various envelopes applied to LFM pulses. A power-raised cosine window with a power p=0.35 and roll-off factor of β=0.15 is shown to maximize the quality of the formed SAR images by improving PSNR by 0.84 dB and SSIM by 0.06 from images formed utilizing a rectangular pulse, on average.

The classification accuracy of a remote sensing image should be assessed before the classification result is used for scientific investigation and policy decision. We proposed an accuracy assessment model based on spatial sampling to reflect region sensitivity of a remote sensing image. The proposed model aims to solve the following problems: (1) what sampling size should be selected for accuracy assessment; (2) where sample points should be distributed in a region; and (3) how to analyze the result of accuracy assessment. This assessment model was proposed based on gray-level co-occurrence matrix (GLCM) and considered both sampling size calculation and sample points distribution during the assessment. The overall accuracy and kappa coefficient derived from this model were very close to the true value derived from the total assessment, suggesting that the assessment accuracy of the model is close to that of total assessment. Compared with the percent sampling model, the model could quantify the relationship between GLCM-correlation parameter and sample size, thereby allowing producer and user to determine sample size according to spatial uniformity and heterogeneity. Compared with the random sampling model, the model could ensure that the sample points are uniformly distributed in the spatial region and proportionally distributed in different types of land cover. Taken together, the proposed model is suitable for the accuracy assessment of the classification result of a remote sensing image.

The assessment of building damage following a natural disaster is a crucial step in determining the impact of the event itself and gauging reconstruction needs. Automatic methods for deriving damage maps from remotely sensed data are preferred, since they are regarded as being rapid and objective. We propose an algorithm for performing unsupervised building segmentation and damage assessment using airborne light detection and ranging (lidar) data. Local surface properties, including normal vectors and curvature, were used along with region growing to segment individual buildings in lidar point clouds. Damaged building candidates were identified based on rooftop inclination angle, and then damage was assessed using planarity and point height metrics. Validation of the building segmentation and damage assessment techniques were performed using airborne lidar data collected after the Haiti earthquake of 2010. Building segmentation and damage assessment accuracies of 93.8% and 78.9%, respectively, were obtained using lidar point clouds and expert damage assessments of 1953 buildings in heavily damaged regions. We believe this research presents an indication of the utility of airborne lidar remote sensing for increasing the efficiency and speed at which emergency response operations are performed.

Remotely sensed imagery is commonly used to map and monitor large land areas based on the ability to detect vegetation stress. Many sensors are available, including both hyper- and multispectral, but have varying costs, convenience, and characteristics. There were two objectives in this study: (1) to compare a hyperspectral sensor to two multispectral sensors with regards to each sensor’s ability to detect vegetation stress indicators in the visible, red edge, near-infrared, and shortwave infrared portions of the spectrum and (2) to determine the ability of coarser-resolution sensors to detect stress indicators in areas, where a finer resolution sensor detected stress indicators. Pairwise agreements between the sensors were ∼80% in each case, but much of this agreement was a function of agreement where stress indicators were absent. Spatial sensitivity analysis supported a conclusion that coarser-resolution sensors were consistently able to detect stress indicators in areas much smaller than their pixel size.

A channel estimation (CE) method based on compressed sensing (CS) is proposed to estimate the sparse and doubly selective (DS) channel for hyperspectral image transmission from unmanned aircraft vehicles to ground stations. The proposed method contains three steps: (1) the priori estimate of the channel by orthogonal matching pursuit (OMP), (2) calculation of the linear minimum mean square error (LMMSE) estimate of the received pilots given the estimated channel, and (3) estimate of the complex amplitudes and Doppler shifts of the channel using the enhanced received pilot data applying a second round of a CS algorithm. The proposed method is named DS-LMMSE-OMP, and its performance is evaluated by simulating transmission of AVIRIS hyperspectral data via the communication channel and assessing their fidelity for the automated analysis after demodulation. The performance of the DS-LMMSE-OMP approach is compared with that of two other state-of-the-art CE methods. The simulation results exhibit up to 8-dB figure of merit in the bit error rate and 50% improvement in the hyperspectral image classification accuracy.

The digital elevation model (DEM) and its derivative attributes are important parameters for evaluating any process using digital terrain analysis. Five freely available global DEM products including Advanced Spaceborne Thermal Emission and Reflection Radiometer-Global Digital Elevation Model version 2 (ASTER GDEM2), Shuttle Radar Topographic Mission version 4.1 (SRTM V4.1), Global Multiresolution Terrain Elevation Data 2010 (GMTED2010), EarthEnv-DEM90, and Global 30 Arc-Second Elevation (GTOPO30) were assessed in this study. The objective of this study was to compare the differences of elevations, slopes, and topographic wetness indices (TWIs) derived from these five DEM products. SRTM V4.1 showed a better accuracy [root mean square error (RMSE)=4.87  m] than ASTER GDEM2 (RMSE=7.08  m) based on ICESat/GLAS (the Ice, Cloud, and land Elevation Satellite/Geoscience Laser Altimeter System) laser altimetry points. ICESat/GLAS data were then selected as the benchmark to rectify the SRTM V4.1 data using the simple kriging (SK) interpolation method. The corrected high-accuracy SRTM V4.1 data (RMSE=1.14  m) were then regarded as the reference data. EarthEnv-DEM90 displayed the best accuracy in the DEM and slope, whereas the TWI accuracy of GMTED2010 was best. The accuracy of topographic attributes was sensitive to the roughness of the terrain. DEM and slope displayed a larger error variance as the elevation increased. DEM was sensitive to the data source and slope was sensitive to the data source and spatial resolution. TWI was influenced by data source and spatial resolution. As the spatial resolution decreased, the differences of topographic attributes tended to decrease.

Suspended particulate matter (SPM) is a key environmental indicator for rivers, estuaries, and coastal waters, which can be calculated from remote sensing reflectance obtained by an airborne or satellite imager. Here, algorithms from prior studies are applied to a dataset of in-situ at surface hyperspectral remote sensing reflectance, collected in three geographic regions representing different water types. These data show the optically inherent exponential nature of the relationship between reflectance and sediment concentration. However, linear models are also shown to provide a reasonable estimate of sediment concentration when utilized with care in similar conditions to those under which the algorithms were developed, particularly at lower SPM values (0 to 20  mg/L). Fifteen published SPM algorithms are tested, returning strong correlations of R2<0.7, and in most cases, R2<0.8. Very low SPM values show weaker correlation with algorithm calculated SPM that is not wavelength dependent. None of the tested algorithms performs well for high SPM values (<30  mg/L), with most algorithms underestimating SPM. A shift toward a smaller number of simple exponential or linear models relating satellite remote sensing reflectance to suspended sediment concentration with regional consideration will greatly aid larger spatiotemporal studies of suspended sediment trends.

The assessment of the dynamic urban structure has been affected by lack of timely and accurate spatial information for a long period, which has hindered the measurements of structural continuity at the macroscale. Defense meteorological satellite program’s operational linescan system (DMSP/OLS) nighttime light (NTL) data provide an ideal source for urban information detection with a long-time span, short-time interval, and wide coverage. In this study, we extracted the physical boundaries of urban clusters from corrected NTL images and quantitatively analyzed the structure of the urban cluster system based on rank-size distribution, spatial metrics, and Mann–Kendall trend test. Two levels of urban cluster systems in the Yangtze River Delta region (YRDR) were examined. We found that (1) in the entire YRDR, the urban cluster system showed a periodic process, with a significant trend of even distribution before 2007 but an unequal growth pattern after 2007, and (2) at the metropolitan level, vast disparities exist in four metropolitan areas for the fluctuations of Pareto’s exponent, the speed of cluster expansion, and the dominance of core cluster. The results suggest that the extracted urban cluster information from NTL data effectively reflect the evolving nature of regional urbanization, which in turn can aid in the planning of cities and help achieve more sustainable regional development.

An objective technique is presented for estimating tropical cyclone (TC) innercore two-dimensional (2-D) surface wind field structure using infrared satellite imagery and machine learning. For a TC with eye, the eye contour is first segmented by a geodesic active contour model, based on which the eye circumference is obtained as the TC eye size. A mathematical model is then established between the eye size and the radius of maximum wind obtained from the past official TC report to derive the 2-D surface wind field within the TC eye. Meanwhile, the composite information about the latitude of TC center, surface maximum wind speed, TC age, and critical wind radii of 34- and 50-kt winds can be combined to build another mathematical model for deriving the innercore wind structure. After that, least squares support vector machine (LSSVM), radial basis function neural network (RBFNN), and linear regression are introduced, respectively, in the two mathematical models, which are then tested with sensitivity experiments on real TC cases. Verification shows that the innercore 2-D surface wind field structure estimated by LSSVM is better than that of RBFNN and linear regression.