Global climate change is expected to result in more frequent and intense drought events, especially during the warm season. In such perspective, it is crucial to assess the forest stands vulnerability to extreme climatic events, such as drought, even for Mediterranean forest tree species, commonly considered resistant to dry spell. To test the capability of multitemporal imagery derived by Sentinel-2 (S2) in detecting the impacts of extreme drought events on forest health assessed as crown dieback, some forest stands in Tuscany (central Italy) were analyzed. Vegetation indices (VIs) and ancillary digital terrain model-derived data have been collected in 118 observational samples distributed along an ecological gradient. VIs detected a reduction of trees of photosynthetic activity in August 2017. S2 data have allowed the observation of the different response strategies of the tree species considered in this study to the extreme climatic event that occurred. The case study presented shows that S2 can be applied for monitoring climate-related processes providing a synthetic overview of forest conditions at regional scale.

This guest editorial introduces the Special Section on Advances in Deep Learning for Hyperspectral Image Analysis and Classification

In multispectral image fusion scenarios, deep learning has been widely applied. However, the fusion performance and image quality are still restricted by inflexible architecture and supervised learning mode. We proposed multispectral image fusion using super-resolution conditional generative adversarial networks (MS-cGANs) based on conditional cGANs, which produces the fused image through the flexible encode-and-decode procedure. In the proposed network, a least square model is extended to solve the gradients vanishing problem in cGANs. Then, to improve the fusion quality, the multiscale features are used to preserve the details. Furthermore, the image resolution is promoted by adding the perceptual loss in object function and injecting the super-resolution structure into a deconvolution procedure. In experimental results, MS-cGANs demonstrates a significant performance in fusing multispectral images and top-ranking image quality compared with the state-of-the-art methods.

The rapid, accurate, and automated extraction of surface water is highly important for conducting reliable and necessary surface water monitoring endeavors. Classification methods commonly exhibit high precision but also have a low degree of automation or narrow scope of application; commonly used water index methods are highly efficient, but they easily mistake other targets with similar spectral characteristics for surface water. Simultaneously achieving precision, efficiency, and automation within a single method is a challenge. To address these problems, we simplify the normalized different water index (NDWI) to a band ratio index and traverse the neighborhood of the extreme in the histogram to determine two peaks and one trough between the peaks in the two-mode method, and we then compare the middle value of the two peaks with the value of the trough to confirm the threshold of the surface water. We use the modified two-mode method to extract Poyang Lake from four Chinese Gaofen (GF)-1 remote sensing images corresponding to different seasons, and then compare the results with those obtained by the NDWI index and the maximization of interclass variance (OTSU) method. The comparison shows that our method has higher and more stable accuracy, especially during the drought period for Poyang Lake. However, polluted water, narrow rivers, bridges, and residential areas along the lake are sometimes mistakenly extracted. Finally, the advantages and prospects of the proposed method are discussed.

Considering the complementarity of Moderate Resolution Imaging Spectroradiometer (MODIS) and Multiangle Imaging Spectroradiometer (MISR) images on temporal and angular resolutions, we propose a fusion method to generate frequent time series MISR images. Thereby, the fusion results can give full play to their respective advantages in the inversion of surface or atmospheric parameters. Based on sparse representation, the proposed method includes two stages: the spectral dictionary-pair training stage and the MISR image prediction stage. In the training stage, we establish a corresponding relationship between the basic MODIS and MISR representation atoms in the spectral domain by learning a dictionary pair from the prior image pairs. In the prediction stage, the MISR images are predicted from the corresponding MODIS images via sparse coding. Experimental results on Baltimore–Washington, DC metropolitan area demonstrate the effectiveness of the proposed method with approximately 7% prediction errors.

Score missing is a common problem involved in exam datasets due to, for example, the absence of students from a test, major-transfer, accidental deletion by operators, etc. Incomplete data tend to bring great inconvenience to analysis, comparison, and evaluation, which may further affect the reliability of the final conclusion and the subsequent decision-making (e.g., adjustment of the teaching plan). Although, in principle, most of the existing data completion methods can be directly used on the exam score dataset, to our best knowledge, there is no systematic evaluation on these methods for the score completion problem. Moreover, the general completion methods cannot effectively employ the special structural information in the score dataset (e.g., the order structure in the students and the relationship between different subjects). Therefore, (1) we conduct a comparative study on the mainstream data completion methods for estimating the missing values in score datasets, (2) we propose an easy-to-implement score completion method by explicitly using informative structures in the dataset, which achieves the best estimation/prediction accuracy with a high efficiency. Beyond the exam scores, we also verify the flexibility of our proposed method in completing remote sensing data.

Convex total variation (TV) regularization models have been widely used in remote sensing image restoration problems; however, these models tend to produce staircase effects. We consider a nonconvex second-order TV regularization model with linear constraints for remote sensing image restoration. To solve the nonconvex second-order TV regularization model, we propose an efficient alternating minimization algorithm based on generalized iterated shrinkage algorithm and alternating direction method of multipliers. Experimental results demonstrate the effectiveness of the proposed model, which can reduce staircase effects while preserving edges. In terms of signal-to-noise ratio and structural similarity index measure, the experimental results show that our proposed model and algorithm can give better performance compared with some state-of-the-art methods.

Classifying land use from postearthquake very high-resolution (VHR) images is challenging due to the complexity of objects in Earth surface after an earthquake. Convolutional neural network (CNN) exhibits satisfied performance in differentiating complex postearthquake objects, thanks to its automatic extraction of high-level features and accurate identification of target geo-objects. Nevertheless, in view of the scale variance of natural objects, the fact that CNN suffers from the fixed receptive field, the reduced feature resolution, and the insufficient training sample has severely contributed to its limitation in the rapid damage mapping. Multiscale segmentation technique is considered as a promising solution as it can generate the homogenous regions and provide the boundary information. Therefore, we propose a combined multiscale segmentation convolutional neural network (CMSCNN) method for postearthquake VHR image classification. First, multiscale training samples are selected based on segments derived from the multiscale segmentation. Then, CNN is directly trained to classify the original image to further produce the preliminary classification maps. To enhance the localization accuracy, the output of CNN is further refined using multiscale segmentations from fine to coarse iteratively to obtain the multiscale classification maps. As a result, the combination strategy is able to capture objects and image context simultaneously. Experimental results show that the proposed CMSCNN method can reflect the multiscale information of complex scenes and obtain satisfied classification results for mapping postearthquake damage using VHR remote sensing images.

The Institute for Atmospheric and Environmental Research at the University of Wuppertal and the Institute of Energy and Climate Research Stratosphere at Research Center Juelich developed a CubeSat payload for atmospheric research. The payload consists of a small interferometer for the observation of airglow near 762 nm. The line intensities of the oxygen A-band are used to derive temperatures in the mesosphere and lower thermosphere region. The temperature data will be used to analyze dynamical wave structures in the atmosphere. The interferometer technology chosen to measure the ro-vibrational structure of the O₂ atmospheric band near 762 nm is a spatial heterodyne interferometer originally proposed by Connes in 1958. It can be designed to deliver extraordinary spectral resolution to resolve individual emission lines. The utilization of a two-dimensional imaging detector allows for recording interferograms at adjacent locations simultaneously. Integrated in a six-unit CubeSat, the instrument is designed for limb sounding of the atmosphere. The agility of a CubeSat will be used to sweep the line-of-sight through specific regions of interest to derive a three-dimensional image of an atmospheric volume using tomographic reconstruction techniques.

Precise radar pulse detection is of great significance for estimating parameters in electronic countermeasure and reconnaissance. An adaptive detection algorithm is proposed, which considers short-time Fourier transforms (STFT), constant false alarm rate (CFAR) in frequency domain, and difference of box (DOB) filter. First, STFT with the Gaussian window is used to acquire the time-frequency spectrum of the radar pulse signal. Second, in order to determine the existence of the pulse, CFAR detector is introduced into the frequency domain to generate an adaptive threshold, and then the rough pulse edges are obtained by mn method. Finally, the data where the rough pulse edges locate are processed by the refined STFT and DOB filter to get the precise pulse edges. The proposed algorithm is processed in the time-frequency domain, which cannot only adapt to low signal-to-noise ratio, but also has a high measurement accuracy. We also draw parallels to the conventional energy-based detection method, the results validate that the proposed algorithm is more robust and effective in practice. Simulations via various noisy input pulse data demonstrate the viability and validity of our proposed algorithm. The algorithm has been implemented in a spaceborne radar receiver.

Rapid urbanization has been an important social and economic phenomenon in the last 50 years. Our study analyzes the spatial–temporal landscape pattern in the National Capital Region (NCR) of Delhi, one of the most rapid urbanization areas in the world. Delhi metropolitan area and its surrounding satellite cities exhibit a soaring rate of landscape pattern change during the last two decades. A set of landscape metrics with supplementary ecological meaning was chosen to study the changes of landscape pattern in NCR. The results indicate that the rapid urbanization has brought enormous landscape changes in NCR, and consequently, substantial impacts on its landscape pattern. The most “active” landscapes are farmland and impervious surface, as the major landscape change (41.46%) is found in the transition from farmland landscape to impervious-surface landscape. Meanwhile, the landscape pattern is fragmented into a more heterogeneous pattern in both farmland and urban landscape with more irregularly shaped patches during urbanization. Our research confirms the effectiveness and applicability of a combination of remote sensing, geographic information systems, and landscape metrics in revealing spatial–temporal of landscape change throughout rapid growth periods.

We propose a methodology for the systematic preparation and processing of interferometric synthetic aperture radar (InSAR) data for monitoring linear transportation infrastructure subject to geohazards. The methodology is applied to two RADARSAT-2 Spotlight synthetic aperture radar datasets, and three case studies in Cornwall, Eastern Ontario, Canada, are examined. An InSAR processing sequence was established and 19 SLA24 and 15 SLA74 images were used to create time-series deformation maps spanning from March 2015 to September 2016. The noise floors were ±1.5 and ±1.0  cm, for the SLA24 and SLA74 datasets, respectively. Phase unwrapping errors, atmospheric path delay, and the limited number of images were identified as the largest contributors to measurement uncertainty, which was of the same order as the ground deformation field. To improve coherence and utility of the radar images for monitoring the effects of geohazards on infrastructure, it is recommended that imagery acquisitions consider the use of small incidence angles with moderate image resolution and 6- to 12-day revisit periods.

The increasing availability of satellite information has improved Earth observation applications globally. However, primary satellite information is not as immediate as desirable. Indeed, besides the geometric and atmospheric limitations, clouds, cloud shadows, and haze generally contaminate optical imagery. Actually, such a contamination is intended as missing information and should be replaced. However, because the most common cloud masking algorithms take advantage by employing thermal images, here the objective is to provide an alternative algorithm suitable for multispectral imagery only. In addition, the work combines a multispectral/multitemporal approach as an effective method to retrieve daytime cloudless and shadow-free optical imagery. Experiment is undertaken upon mid- to low-spatial resolution data from Landsat 5 TM and Landsat 8 OLI, each for a different scene. A multitemporal stack, for the same image scene, is employed to retrieve a composite uncontaminated image over 1 year. The approach relies on a clouds and cloud shadows masking step, based on spectral features, a band-by-band multitemporal effect adjustment to avoid significant seasonal variations, and a data reconstruction phase based on automatic selection of the most suitable pixels from the stack. Results have been compared with a recognized masking algorithm approach and tested with uncontaminated image samples for the same scene. Accuracy and spectral features of the results provide high consistency.

In order to make full use of local neighborhood information for high-resolution remote sensing images, this study combined iterative slow feature analysis (ISFA) and stacked denoising autoencoder (SDAE) to improve the change detection precision. First, this approach introduced ISFA for initial change detection in an unsupervised way, which enlarged the separability of changed and unchanged areas. Then, by setting different membership degrees, the changed and unchanged samples were obtained through fuzzy-means clustering. Finally, the change model was built by SDAE to represent the local neighborhood features deeply, and the change detection result can be obtained after all the samples were fed into the model. Experiments were performed on three real datasets, and the results validated the effectiveness and superiority of the proposed approach.

Polarimetric synthetic aperture radar (PolSAR) images are disturbed by an inherent noise having multiplicative properties called “speckle.” This noise is undesirable, and its treatment is difficult. To reduce the speckle, a polarimetric filtering is necessary to improve the image quality. The purpose of PolSAR filtering is to use the polarimetric information in the different channels to develop an efficient algorithm adapted to this data type, to reduce well the speckle and preserve the contained information. We present the PolSAR wavelet filtering applying the stationary wavelet transform: filtering by multiscale edge detection with two improvement techniques of wavelet coefficients, filtering by wavelet thresholding using the hard and soft thresholding and their two enhanced versions. Our contribution is based on the adaptation of wavelet thresholding to PolSAR data and on improvement techniques to filter polarimetric covariance or coherency matrix elements and span. The methods are applied to the fully polarimetric RADARSAT-2 images acquired over Algiers, Algeria (C-band), to the three polarimetric E-SAR images acquired on Oberpfaffenhofen area located in Munich, Germany (P-band), and to the simulated PolSAR images (L-band). We evaluate the performance of each filter based on the following criteria: smoothing homogeneous areas, preserving contours, and polarimetric information. Experimental results and a comparative study are included.

Nowadays, many of the world’s large cities are faced with the issue of land scarcity for construction due to the increasing growth of urbanization, as well as the economic downturn for exploiting lands and properties, and city officials have come up with the idea of optimal management of real estate in order to cope with these problems. The purpose of our study is to reconstruct three-dimensional (3-D) building cadastre models (3DBCMs) with an approach to improve the state of land administration in Tehran metropolis. Our study is being implemented and evaluated in three stages. The first stage involves collecting aerial images. The interior and exterior orientation parameters are preprepared in this step. The second stage involves automatic interpretation and extraction of buildings from aerial images by providing a method of interpretation called fully automatic interpretation with deep learning (FAIDL). The third stage involves 3-D building modeling and evaluating the effect of FAIDL method on the automatic interpretation of images. The results showed that the 3-D models of building have a better geometric accuracy compare to 60 cm, which are produced with the proposed algorithm.

We have previously shown the advantage of using neural network (NN) inversion algorithms over other ocean color (OC) algorithms in Visible Infrared Imaging Radiometer Suite satellite retrievals of Karenia brevis (KB) in the west Florida shelf (WFS). We now extend NN retrievals well beyond the WFS, to include both complex coastal and open ocean waters along the Florida and Atlantic coasts with a large dynamic range of chlorophyll-a values. Most importantly, we add in situ radiometric measurements (which in contrast to satellite retrievals, are invulnerable to atmospheric transmission correction errors) as inputs to retrieval algorithms, permitting algorithm comparisons for in situ and simultaneous colocated satellite retrievals against sample measurements. Results unequivocally demonstrate the intrinsic efficacy and unfettered applicability of NN algorithms in widely varying waters beyond the WFS. Furthermore, they show that avoiding deep blue bands in retrieval algorithms significantly improves accuracies. Likely, rationales are that longer wavelengths (used with NN) are less vulnerable to atmospheric transmission correction errors and to spectral interference by colored dissolved organic matter and nonalgal particles in more complex waters than deeper blue wavelengths (used with other algorithms), thereby arguing for development of OC algorithms using longer wavelengths. Finally, quantitative analysis of temporal, intrapixel, and sample depth variabilities highlights their important impact on retrieval accuracies.

As a fundamental prerequisite for a variety of location-based services, indoor location information has received increasing attention in recent years. Under the line-of-sight condition, the positioning accuracy of the indoor positioning technology based on ultrawideband (UWB) is acceptable for many applications, but under the non-line-of-sight condition, it degrades dramatically. The positioning accuracy can be significantly improved by the fusion of inertial measurement units and UWB sensors based on the extended Kalman filter (EKF) algorithm. However, when UWB measurements are affected by large non-Gaussian noise, the assumption of the EKF algorithm that observations are subject to Gaussian distribution for noise is invalid. Although the non-Gaussian noise can be handled by the robust EKF algorithm, this algorithm only uses the prior information to judge the reliability of the observations, and the positioning result is not stable when the number of beacons is small. To solve this problem, a method for successive updating of the covariance and posterior state of the observations in iterations based on an iterated extended Kalman filter (IEKF) is proposed. The marginal distribution of the posterior distribution is constructed and iteratively optimized, inhibiting the effect of non-Gaussian noise on UWB under a complex environment. The positioning results of the proposed method, the standard EKF algorithm, and the robust EKF algorithm, using different numbers of beacons, are compared. The results show that the positioning accuracy of the proposed algorithm is the highest under all scenarios. The proposed algorithm shows the smallest decrease in accuracy and presents the most stable positioning when the number of beacons is small, which is a common situation in practical applications.

As a crucial parameter in land surface systems, soil moisture plays an important role in surface energy balance studies, environmental detection, and global climate change research. Remotely sensed data have been used for estimating soil moisture through different approaches, which has resulted in many achievements. Previous studies showed that the land surface temperature (LST) vegetation index method (LST-VI method) can obtain surface soil moisture with remote sensing sources, and it is relatively simple and easy to operate at a regional scale. However, one thorny difficulty is the dry edge determination from the LST-VI feature space. In this study, a remote sensing method is proposed to determine the theoretical dry edge from the LST-VI scatter plots, which do not require any ground measured auxiliary data. Based on the surface energy balance principle, this method derived the maximum LSTs for bare soil and full vegetation cover using MODIS products. The air temperature is parameterized by the LST using a semiempirical formula as the theoretical wet edge. The estimated soil moisture is validated by in situ measurements at a comprehensive weather station of Yucheng. The coefficient of determination is ∼0.60, and the root mean square error is about 0.08  m³  /  m³. The relevant key parameters in determining the dry edge are also validated from the meteorological observation. The air temperature and net surface shortwave radiation flux all reach a very high level, with an RMSE of 3.75 k and 49.3  W m^−  2, respectively. The results demonstrated that the proposed method can derive the accurate dry edge to estimate soil moisture from the remote sensing data, which will provide great help for future studies of soil moisture estimation using remote sensing techniques.

Due to great advantages in deep features representation and classification for image data, deep learning is becoming increasingly popular for change detection (CD) in the remote-sensing community. An unsupervised CD method is proposed by combining deep features representation, saliency detection, and convolutional neural network (CNN). First, bitemporal images are fed into the pretrained CNN model for deep features extraction and difference image generation. Second, multiscale saliency detection is adopted to implement the uncertainty analysis for the difference image, where image pixels can be categorized into three classes: changed, unchanged, and uncertain. Then, a flexible CNN model is constructed and trained using the interested changed and unchanged pixels, and the change type of the uncertain pixels can be determined by the CNN model. Finally, object-based refinement and multiscale fusion strategies are utilized to generate the final change map. The effectiveness and reliability of our CD method are verified on three very high-resolution datasets, and the experimental results show that our proposed approach outperforms the other state-of-the-art CD methods in terms of five quantitative metrics.

For Earth-observing satellites in low-Earth orbit, radiometric calibration of the sensors on-orbit is critical for maintaining consistent Earth-view (EV) retrievals as the mission progresses. Many of these satellite instruments use on-board calibration targets, EV sites, and observations of celestial targets in order to perform the sensor characterization. Among the celestial targets, the Moon is widely used across a range of Earth-observing instruments in order to perform radiometric calibration, spatial characterization, and sensor intercomparison. Since many of these instruments use satellite maneuvers in order to bring the Moon into view at a desired time, calculating the time and geometric parameters of the observations is vital for mission planning purposes. We develop a simple tool for planning such observations of the Moon and other celestial targets for instruments in low-Earth orbit based on the SPICE toolkit developed by the Navigation and Ancillary Information Facility at NASA. Given a set of satellite orbital data along with a definition of the instrument coordinates, the tool is designed to provide the timing of observations for an arbitrary view-port direction and a maneuver along an arbitrary axis relative to the spacecraft. The tool can be tested versus known lunar observations for the Aqua and Terra moderate resolution imaging spectroradiometer and the Suomi-NPP and NOAA-20 visible infrared imaging radiometer suite instruments for both roll and pitch maneuvers. We also perform simulations of lunar observations for different instrument configurations, orbits, and maneuver types in order to analyze the change in the potential lunar observations. Finally, we show a simple extension of the tool which can be used for identifying planet and star observations.

With the rapid development of urban areas, construction areas are constantly appearing. Those changed areas require timely monitoring to provide up-to-date information for urban planning and mapping. As a result, it is a challenge to develop an effective change detection technique. In this work, a method for detecting building changes from multitemporal high-resolution aerial images is proposed. Different from traditional methods, which usually depict building changes in the color domain (e.g., using pixel values or its variants as features), this work focuses on analyzing building changes in the spatial domain. Moreover, contextual relations are explored as well, in order to achieve a robust detection result. In detail, corners are first extracted from the image and an irregular Markov random field model is then constructed based on them. Energy terms in the model are appropriately designed for describing the geometric characteristics of the building. Change detection is treated as a classification process, so that the optimal solution indicates corners belonging to changed buildings. Finally, changed areas are illustrated by linking preserved corners followed by postprocessing steps. Experimental results demonstrate the capabilities of the proposed method for change detection.

We use X-band Cosmo-SkyMed InSAR data to highlight several subsidence phenomena resting on some railway and road infrastructures in Lombardia region, Northern Italy, mainly induced by anthropogenic activities. We show eight case studies, namely “Como,” Erba, Oggiono, Valmadrera, Olginate, Verduggio, Melzo, and San Giuliano M., where we detect local subsidence effects affecting several railway and highway lines with deformation rates of about 5 mm/year. The geological features of this part of Italy and the large presence of industrial areas in the surrounding of Milano, Lecco, and Como cities lead to such phenomena. The stability and security of the nested road and railway network could be affected by these surface deformation fields. To guarantee the safety of people, continuous maintenance of the condition of railways and roads together with the monitoring of the conditions of the lands on which they rest on should be done.

Ground-truthing results are presented for a new 1-km air temperature product downscaled for New York City (NYC) from ∼12  km North American Land Data Assimilation System (NLDAS) air temperature data using 1 km moderate resolution imaging spectroradiometer surface temperature data. The downscaled product was compared against a unique highly spatially resolved ground-level ambient air temperature dataset collected through the New York City Community Air Survey (NYCCAS), a neighborhood level air pollution and temperature monitoring network, for the years 2009 and 2010. This work focuses on the spatial variation in daily minimum temperatures within the five counties that comprise NYC (∼784  km²). Overall, the downscaled daily minimum temperature was well correlated with ground station data, with NYCCAS minimum temperatures being slightly higher. Minimum temperature R² values were 0.9 and 0.92, and mean absolute errors were 0.69°C and 0.86°C for years 2009 and 2010, respectively. The smallest differences between NYCCAS and the downscaled data were seen at lower temperatures, in less densely urbanized areas, and in areas with higher vegetative cover, suggesting systematic bias in the downscaled data related to land-use. The 1-km dataset discerned neighborhood level temperature differences in high-density urban situations with heterogeneous land cover.

A hyperspectral (HS) imager is more effective than a multispectral (MS) imager in mineral discrimination, but spatial coverage of HS images is limited in comparison to that of MS images. Thus Kruse and Perry have proposed a method that uses coincident HS imaging and MS imaging data to extend mineral mapping to larger areas. We propose a method modified from the Kruse and Perry’s (K&P) method. Though the K&P method derives the MS-based endmember spectra by weighting the HS-based endmember spectra with the response functions of the MS sensor bands, the proposed method obtains the MS-based endmember spectra from surface reflectance spectra of the MS pixels at the same positions with the HS pixels selected as the HS-based endmembers in the overlapping area. The validation study using airborne visible/infrared imaging spectrometer and advanced spaceborne thermal emission and reflection radiometer images over Cuprite and Goldfield areas, Nevada, USA, demonstrates that the proposed method is more robust against spectral inconsistency between the HS- and the MS-images caused by calibration and/or atmospheric correction errors than the K&P method, though the proposed method is more sensitive to co-registration errors between the HS- and the MS-images than the K&P method.

Remote sensing technique often analyzes the thermal characteristics of any area. Our study focuses on estimating land surface temperature (LST) of Raipur City, emphasizing the urban heat island (UHI) and non-UHI inside the city boundary and the relationships of LST with four spectral indices (normalized difference vegetation index, normalized difference water index, normalized difference built-up index, and normalized multiband drought index). Mono-window algorithm is used as LST retrieval method on Landsat 8 Operational Land Imager (OLI) and Thermal Infrared Sensor (TIRS) data, which needs spectral radiance and emissivity of TIRS bands. The entire study is performed on 11 multidate Landsat 8 OLI and TIRS images taken from four different seasons; premonsoon, monsoon, postmonsoon, and winter, in a single-year time period. The Landsat 8 data derived LST is validated significantly with Moderate Resolution Imaging Spectroradiometer (MOD11A1) data. The results show that the UHI zones are mainly developed along the northern and southern portions of the city. The common area of UHI for four different seasons is developed mainly in the northwestern parts of the city, and the value of LST in the common UHI area varies from 26.45°C to 36.51°C. Moreover, the strongest regression between LST and these spectral indices is observed in monsoon and postmonsoon seasons, whereas winter and premonsoon seasons revealed comparatively weak regression. The results also indicate that landscape heterogeneity reduces the reliability of the regression between LST with these spectral indices.

Soil moisture (SM) at three depths (15, 25, and 30 cm), derived from the optical trapezoidal model (OPTRAM), was used for multiyear, multisite monitoring of agricultural droughts over two agricultural crops (Maize and Soybean) in southern Mozambique. The OPTRAM was implemented using satellite data from Sentinel-2 and was validated against field SM assessed by gravimetric methods and by Watermark Sensors in sandy-soils with very low water holding capacity (0.13  cm³  /  cm³). The OPTRAM model estimated the SM at 15 and 25 cm yielding a R²  ≥  0.79 and RMSE  ≤  0.030  cm³  /  cm³. The OPTRAM-derived SM was successfully used as input to compute and map the soil water deficit index, an indicator of agricultural drought. The results indicate that OPTRAM can provide useful information to improve water productivity in cropland under the specific conditions of Mozambique agricultural systems and for early warning systems development.

The potential of synthetic aperture radar (SAR) interferometry was shown to study the compaction of the aquifer system in Darab plain, Iran. In so doing, two different datasets, including Envisat advanced SAR (ASAR) spanning 2010 and Sentinel-1A spanning 2016 to 2017, were applied in small baseline subset time series analysis. To estimate the subsidence in the time period for which there is no SAR data available, i.e., 2010 to 2016, the time series analysis results separately obtained from the two datasets were to be integrated using an appropriate model, which should have been fitted to both sets of results. However, as both deformation time series results were calculated taking into account a distinct temporal reference, fitting the model was not a straightforward task. Accordingly, the main attempt was to find the subsidence value corresponding to the temporal reference of Sentinel-1A time series with respect to that of Envisat ASAR. This shift value was optimally determined using a genetic algorithm so as to minimize the misfit between the model and the deformation time series corresponding to the entire period. The average value of the root mean square error estimated as the misfit between the model and the calculated time series at all pixels is 0.011 m, which is an indication of the high performance of the proposed method for modeling the deformation time series. The integration results were further used to derive the stress–strain relationships to study the storage properties of the aquifer system. The fact that the strain linearly increases along with the decrease in water level in most piezometric wells indicates that the subsidence is highly correlated with groundwater exploitation.

The Suomi National Polar-orbiting Partnership visible infrared imaging radiometer suite instrument has successfully operated since its launch in October 2011. Stray-light contamination is much larger than prelaunch expectations, and it causes a major decrease in quality of the day-night band night imagery when the spacecraft is crossing the Northern or Southern day-night terminators. The stray light can be operationally estimated using Earth-view data that are measured over dark surfaces during the new moon each month. More than 7 years of nighttime images have demonstrated that the stray-light contamination mainly depends on the Earth–Sun–spacecraft geometry, so its intensity is generally estimated as a function of the satellite zenith angle. In practice, stray-light contamination is also detector- and scan-angle-dependent. Previous methods of stray-light prediction generally rely on using the known stray light level from the same month in the previous year, when the Earth–Sun–spacecraft geometries had been similar. We propose a new method to predict stray-light contamination. The Kullback–Leibler similarity metric is used as a method to combine data from multiple years with appropriate adjustments for degradation and geometry drifts in order to calculate a fused stray-light contamination correction. The new method provides an improved prediction of stray-light contamination compared to the existing methods and may be considered for future use in the real-time NASA Level-1B products.

Porosity is a fundamental characteristic of naturally occurring sand-textured soils, commonly referred to as natural sands, found in a wide range of landscapes, from beaches to dune fields. As a primary determinant of the density and permeability of sediments, it represents a pivotal element in geophysical studies involving basin modeling and the optical dating of sand deposits formed in areas subjected to erosion like coasts and deltas. It is also of interest for geoaccoustics and geochemical research on sediment transport and water diffusion properties of these deposits, as well as for agricultural and ecological investigations on the germination of light-sensitive seeds and the photochemical transformation of substances (e.g., pesticides) that may be present in these soils. Despite the importance of these applications, however, the remote estimation of porosity and the quantification of its effects on the light penetration profiles of natural sands remain elusive tasks. In this work, we tackle one of the major obstacles in this interdisciplinary area of research, namely the relative scarcity of experimental information due to technical constraints associated with traditional laboratory procedures. More specifically, we systematically examine the impact of porosity variations on the reflectance and transmittance of natural sands (in the 400 to 1000 nm region of the light spectrum) through controlled in silico experiments supported by measured data. Our findings are expected to strengthen the knowledge basis required for advances in this area and contribute to the development of technologies aimed at the effective monitoring and prediction of environmentally triggered changes affecting these soils.

The spatial heterogeneity of urban vegetation obtained from discrete land cover classes is sensitive to classification errors and can result in a substantial loss of information, due to the degradation of continuous quantitative information. Although there is an increasing ecological need to use continuous methods to understand spatial heterogeneity and vegetation fragmentation, they remain unexplored. Since local indicators of spatial association (LISA) can capture important spatial patterns of clustering and dispersion at a local scale, it can capture important ecological patterns and process of vegetation fragmentation. This work examines the utility of LISA which allows exploration of local patterns in spatial data in identifying high (hot spots) and low (cold spots) spatial clusters of vegetation patches and fragmentation patterns in Harare metropolitan city in Zimbabwe. The LISA indices of Getis-Ord Gi* and local Moran’s I are computed both on continuous normalized difference vegetation index and discrete land cover data of vegetation and nonvegetation of Sentinel 2016 and 2018. Local spatial clustering patterns are identified with Z-score values that indicate the significance of each statistic. High positive Z-scores are located in the large core, undisturbed, and homogeneous vegetation. Negative Z-scores are located in more dispersed and highly fragmented vegetation. The results suggest that there is a strong tendency for large core, undisturbed, and homogeneous vegetation patches to be spatially clustered and for small, isolated and sparse vegetation patches to be dispersed. The highly fragmented vegetation patches are located in the heavily urbanized part of the city. Overall, findings of this study underscore the relevance of the spatially explicit method of LISA as a valuable source of spatial information for the assessment of local spatial clustering and dispersion of urban vegetation patches. This can be used to develop policies that support effective conservation and restoration strategies.

Viewing geometry is one of the most important factors to consider when water bodies are observed from satellite sensors with large field of view. We examine the directional and angular effects on the reflectance of waters with different concentrations of total suspended solids (TSSs). In the laboratory, we measure the reflectance in five view zenith angles (VZAs) and eight view azimuth angles (VAAs) for optically shallow waters having four concentrations of TSSs. Seven empirical models to estimate TSSs based only on the reflectance of the red band (∼660  nm) are evaluated. In addition, we analyze Moderate Resolution Imaging Spectroradiometer (MODIS)/Terra reflectance measured in 13 consecutive satellite overpasses. The results show that the reflectance of the inland-like water is affected by data acquisition geometry. The best wavelength to estimate TSS is 625 nm for most VZAs and VAAs. The lowest correlations between reflectance and TSS are observed at extreme viewing with the anisotropy decreasing with increasing concentrations of TSSs. Directional and angular effects are also observed for MODIS (acquired and simulated data) with TSS underestimates observed close to the orthogonal plane for all VZAs, and TSS overestimates observed in the principal scattering plane in the forward scattering direction. More anisotropic waters are observed for VZA greater than ±30  deg. Results highlight the need for correcting MODIS data for bidirectional effects in inland water studies.

Currently, analyzing satellite images requires an unsustainable amount of manual labor. Semiautomatic solutions for land-cover classification of satellite images entail the incorporation of expert knowledge. To increase the scalability of the built solutions, methods that automate image processing and analysis pipelines are required. Recently, deep learning (DL) models have been applied to challenging vision problems with great success. We expect that the use of DL models will soon outperform shallow networks and other classification algorithms, as recently achieved in multiple domains. Here, we consider the task of land-cover classification of satellite images. This seems particularly appropriate for deep classifiers due to the combined high dimensionality of the data with the presence of compositional dependencies between pixels, which can be used to characterize a particular class. We develop a pipeline for analyzing satellite images using a deep convolutional neural network for practical applications. We present its successful application for land-cover classification, where it achieves 86% classification accuracy on unseen raw images.

We develop two endmembers abundance estimators based on a genetic algorithm (GA) and a generalized pattern search algorithm. This development aims to estimate more accurate abundances of endmembers in cases of brightness and shading noises, which is an issue in other endmember abundance estimation methods based on the least square method, such that the estimators depend on spectral shape similarity as a matching criterion between spectral signatures, because shape similarity methods are independent on the spectral reflectance amplitude and not sensitive to brightness noise effects on it. The strategy used for unmixing problem analysis is based on the popular linear spectral mixture model. GA is used as a heuristic optimization technique, and generalized pattern search is used as a direct search algorithm. Spectral angle mapper, spectral information divergence methods, and the combination of both of these are used as objective functions for optimization techniques. All experiments have been performed on the proposed estimators and the traditional fully constrained least squares method as a state-of-the-art method. Experiments have been applied on simulated multispectral and hyperspectral datasets with different noise conditions. In addition, a hyperspectral real dataset from a cuprite region (Las Vegas, Nevada, USA) was used for testing performance. The experimental results show that proposed estimators are better than fully contained least squares method across all experiments and especially in cases of noisy datasets.

We present a hierarchical classification framework for automated detection and mapping of spatial patterns of agricultural performance using satellite-based Earth observation data exemplified for the Aral Sea Basin (ASB) in Central Asia. The core element of the framework is the derivation of a composite agricultural performance index which is composed of different subindicators taking into account cropping intensity, crop diversity, crop rotations, fallow land frequency, land utilization, water use efficiency, and water availability. We derive these subindicators from net primary productivity and evapotranspiration data obtained from the MODIS sensor on board the Terra satellite during the observation period from 2000 to 2016, as well as from cropland maps created through multiannual classification of normalized difference vegetation index (NDVI). We classified pixel-based NDVI time series covering more than 8  ×  ¹⁰⁶  ha of irrigated cropland based on a hierarchical approach concatenating unsupervised and supervised classification techniques to automatically generate and refine training labels, which are then used to train a decision fusion classifier, achieving an average overall accuracy of 78%. The results give unprecedented insights into spatial patterns of agricultural performance in the ASB. The proposed method is transferable and applicable for global-scale mapping, and the results of this remote sensing-aided assessment can provide important information for regional agricultural planning purposes.

We aim to classify every pixel of a hyperspectral image. Toward this goal, we first decompose the image data. The three-dimensional (3-D) tensor of an image cube is decomposed to the spectral signatures and abundance matrix using non-negative tensor factorization (NTF) methods. In contrast to the matrix factorization where the pixels’ spectra are stacked in columns of the data matrix, the NTF techniques preserve the spatial structure of the image data. Therefore, the obtained abundance maps provide discriminant spatial features for classification. Morphological attribute profiles are also computed for abundance maps and their effect on the classification performance is studied. Using the original spectral image and the obtained spatial features of the abundance matrices, we construct a composite kernel framework. We apply a multinomial logistic regression classifier to the kernels. Experiments show that, in terms of the classification accuracy, the proposed feature sets acquired through NTF techniques can lead to a better performance compared to the principal component analysis and non-negative matrix factorization feature sets.

An effective onboard cloud detection method in small satellites will greatly improve the downlink data transmission efficiency and reduce the platform memory cost. A methodology combining a convolutional neural network and wavelet image compression is proposed to explore the possibility of onboard cloud detection. A lightweight neural network based on U-net architecture is established and evaluated. The red, green, blue, and infrared waveband images from the Landsat-8 dataset are trained and tested to estimate the performance of the mythology. Then a LeGall-5/3 wavelet transform is applied on the dataset to accelerate the neural network and improve the feasibility of the onboard implementation. The experiment results on advanced RISC machines-based embedded platform illustrate that by taking advantage of a mature image compression system in small satellites; the time cost and peak memory cost required by the neural network will be reduced significantly while the segment accuracy is only slightly decreased. The proposed method provides a good possibility of onboard cloud detection for small satellites.

Ground penetrating radar (GPR) is a powerful technology for detection and identification of buried explosives, especially with little or no metal content. However, subsurface clutter and soil distortions increase false alarm rates of current GPR-based landmine detection and identification methods. Most existing algorithms use shape-based, image-based, and physics-based techniques. Analysis of these techniques indicates that each type of algorithm has a different perspective to solve the landmine detection and identification problem. Therefore, one type of method has stronger and weaker points with respect to the other types of algorithms. To reduce false alarm rates of the current GPR-based landmine detection and identification methods, we propose a combined feature utilizing both physics-based and image-based techniques. Combined features are classified with a support vector machine classifier. The proposed algorithm is tested on a simulated data set that contained more than 500 innocuous object signatures and 400 landmine signatures, over half of which are completely nonmetal. The results presented indicate that the proposed method has significant performance benefits for landmine detection and identification in GPR data.

Endmember extraction is the key step in the mixed pixel decomposition for hyperspectral images. In view of the larger Markov property of endmember error in the sequence endmember extraction algorithm, which affects the endmember extraction accuracy, we propose an endmember extraction algorithm with three endmembers as a group based on Gram–Schmidt orthogonalization. According to the convex geometry theory, the spectral characteristic and the geometrical property of simplex in feature space have been analyzed, and the idea of group endmember extraction was introduced to reduce the Markov property of the endmember error, improving the endmember extraction accuracy accordingly. The orthogonal vector was searched by Gram–Schmidt orthogonalization and the image was projected to the orthogonal vector, so as to eliminate the effect of the extracted endmembers. The energy function was used as a measure index of the similarity for spectral vectors of different ground objects, and the measure index was used to determine the endmember. The algorithm was verified by using simulation data and real data. The experimental results indicated that the proposed algorithm may extract endmember automatically, and the corresponding endmember extraction accuracy was better than other algorithms.

The mechanism of speckle noise in synthetic aperture radar (SAR) images and its characteristics are analyzed. Combining the advantages of the traditional bilateral filter (BF) and alpha-trimmed median filter, a truncated-statistics-based bilateral filter (TS-BF) in SAR imagery is proposed. The despeckling method is based on the BF methodology, where the similarities of gray levels and spatial location of the neighboring pixels are exploited. However, traditional BF is not effective to reduce the strong speckle, which is often presented as impulse noise. The proposed TS-BF filtering method designs an adaptive truncation method to properly select the samples in the local reference window, where the mean and standard deviation of all the samples are estimated, and the background types of the current pixel-for-filtering are categorized. Finally, the samples of the local reference window are truncated with different levels according to different background types, and BF is applied using the truncated samples. TS-BF can effectively preserve the edge and texture information of the image while smoothing the speckle noise; it has a great application value. The experimental results show the effectiveness of the proposed algorithm through subjective and objective analyses.

A probabilistic topic model (PTM) combined with the bag-of-visual-words model is a common method to bridge the so-called “semantic gap” problem in remote-sensing image classification research. Owing to the inherent shortcomings of PTMs, such as time consumption and failures to consider a spatial arrangement of various objects, we introduce a natural language processing document-to-vector (Doc2Vec) model, to capture the high-level semantic information of the images, instead of a PTM. The model characterizes words and documents as dense, low-dimensional vectors and implements a simplified, shallow neural network to train a language model and word vectors. It is expected to mine semantic information of remote-sensing images from a new perspective. We also improve the low-level feature quality by using feature-specific sampling methods. Two high-spatial resolution remote-sensing image datasets, UC Merced and RSSCN7, are employed to conduct a scene classification experiment to discuss the performance of the Doc2Vec model. The experimental results show that the Doc2Vec model is highly efficient in mining semantic information of the images, compared with the state-of-the-art methods.

Pan-sharpening is an indispensable technology for remote sensing that aims to combine low-resolution multispectral images and high-resolution panchromatic images to create a multispectral image with high resolution. However, pan-sharpening approaches often encounter spectral distortion and detail distortion issues. In order to overcome the drawbacks of pan-sharpening methodologies, we propose an end-to-end pan-sharpening model consisting of an effective generative adversarial network architecture equipped with spatial feature transform layers that generate spatial detail features under spectral feature constraints. Through a large number of quantitative and visual assessments, we demonstrate that the proposed method achieves superior performance to other state-of-the-art methods.

Hyperspectral imaging is a powerful technology that is plagued by large dimensionality. Our study explores a way to combat that hindrance via noncontiguous and contiguous (simpler to realize sensor) band grouping for dimensionality reduction. Our approach is different in the respect that it is flexible and it follows a well-studied process of visual clustering in high-dimensional spaces. Specifically, we extend the improved visual assessment of cluster tendency and clustering in ordered dissimilarity data unsupervised clustering algorithms for supervised hyperspectral learning. In addition, we propose a way to extract diverse features via the use of different proximity metrics (ways to measure the similarity between bands) and kernel functions. The discovered features are fused with ℓ_∞-norm multiple kernel learning. Experiments are conducted on two benchmark data sets and our results are compared to related work. These data sets indicate that contiguous or not is application specific, but heterogeneous features and kernels usually lead to performance gain.

The algorithm based on a clustered multitask network is proposed to solve spectral unmixing problem in hyperspectral imagery. In the proposed algorithm, the clustered network is employed. Each pixel in the hyperspectral image is considered as a node in this network. The nodes in the network are clustered using the fuzzy c-means clustering method. Diffusion least mean square strategy has been used to optimize the proposed cost function. To evaluate the proposed method, experiments are conducted on synthetic and real datasets. Simulation results based on spectral angle distance, abundance angle distance, and reconstruction error metrics illustrate the advantage of the proposed algorithm, compared with other methods.

A level set model is presented for water region segmentation in synthetic aperture radar (SAR) images. We formulate the segmentation problem within a global energy minimization framework. First, the background and foreground regions in SAR images are modeled as G⁰ distributions. They are then used to construct the energy functional for the desired regions. To avoid the local minimum problem, the energy functional is transferred into a strictly convex model that guarantees the existence of the global minimum. During the iterative process, a sinusoidal signed pressure force (SPF) function is applied to efficiently locate weak or blurred edges in the heterogeneous regions. Finally, a Gaussian convolution is used to equivalently substitute the Laplacian of the level set function in the evolution equation, which omits the reinitialization at each iteration. Since based on the stationary global minimum, the presented model can accurately detect inside edges, regardless of the position and shape of the initial contour. Furthermore, because the SPF function can enhance the acquisition ability to the target contour, the internal and external motions of the curve can be accelerated. Thus, the convergence speed of the curve can be improved significantly. The experimental results based on the simulated and real SAR data demonstrate the effectiveness of our method.

In port surveillance, monitoring based on satellite video is a valuable supplement to a ground monitoring system because of its wide monitoring range. Therefore, automatic ship detection and tracking based on satellite video is an important research field. However, because of the small size of ships without texture and the interference of sea noise, it is also a challenging subject. An approach of automatic detection and tracking moving ships of different sizes using satellite video is presented. First, motion compensation between two frames is realized. Then, saliency maps of multiscale differential image are combined to create dynamic multiscale saliency map (DMSM), which is more suitable for the detection of ships of different sizes. Third, candidate motion regions are segmented from DMSM, and moving ships can be detected after the false alarms are removed based on the surrounding contrast. Fourth, important elements such as centroid distance, area ratio, and histogram distance from moving ships are used to perform ship matching. Finally, ship association and tracking are realized by using the intermediate frame in every three adjacent frames. Experimental results on satellite sequences indicate that our method can effectively detect and track ships and obtain the target track, which is superior in terms of the defined recall and precision compared with other classical target tracking methods.

Hyperspectral images (HSIs) contain spectral information on the order of hundreds of different wavelengths, providing information beyond the visible range. Such spectral sensitivity is often used for the classification of objects of interest within a spatial scene in fields, such as studies of the atmosphere, vegetation and agriculture, and coastal environments. The classification task involves the processing of high-dimensional data which fuels the need for efficient algorithms that better use computational resources. Classification algorithms based on sparse representation classification perform classification with high accuracy by incorporating all the relevant information of a given scene in a sparse domain. However, such an approach requires solving a computationally expensive optimization problem with time complexity Ω  (  n²  )  . We propose a method that approximates the least squares solution of the sparse representation classification problem for HSIs using the Moore–Penrose pseudoinverse. The resulting time complexity of this approach reduces to O  (  n²  )  . The impact on the classification accuracy and execution time is compared to the state-of-the-art methods for three varied datasets. Our experimental results show that it is possible to obtain comparable classification performance current methods, with as much as 82% of a reduction in execution time, opening the door for the adoption of this technology in scenarios where classification of high-dimensional data is time critical.

Anomaly detection (AD) is an important technique for hyperspectral image processing and analysis. Typically, it is accomplished by extracting knowledge from the background and distinguishing anomalies and background using the difference between them. However, it is almost impossible to obtain “pure” background to achieve an ideal detection because of anomaly contamination. The low-rank and sparse matrix decomposition (LRaSMD) technique has been proved to have the potential to solve the aforementioned problem. But the accuracy and time consumption need to be further improved. Thus we propose a local hyperspectral AD method based on LRaSMD with an optimization algorithm for better performance. The LRaSMD technique is first implemented with semisoft Go decomposition (GoDec) rather than GoDec to quickly and accurately set the background apart from the anomalies. Then the low-rank prior knowledge of the background is fully explored to compute the background statistics. After that, the local Mahalanobis distance of pixels is calculated with the sliding dual-window strategy to detect the probable anomalies. The proposed method is validated using four real hyperspectral data sets with ground-truth information. Our experimental results indicate that the proposed method achieves better detection performance as compared with the comparison algorithms.

Object detection from hyperspectral images (HSIs) is an important issue but encounters a critical challenge that results in poor detection due to the variation of the detection object spectrum. Especially when the detection object area is large and widely distributed in HSIs, such spectral variability becomes more serious. Spectral variability can make false detection and leak detection in object detection very serious. The constrained energy minimization (CEM) algorithm is a classical object detection algorithm that only needs the object prior spectrum to achieve object detection, but the spectral variability will have a detrimental effect on the detection results of the CEM algorithm. To address the above problems, we propose a multiobject subspace projection sample weighted CEM (MSPSW-CEM) algorithm. The proposed method has the following capabilities: (1) it constructs object subspaces and detectors using multiple prior spectra of the detection object under spectral variability conditions and (2) it utilizes the subspace projection theory to weight the pixel spectra, so that the detector can better suppress the background information and highlight the object information. Extensive experiments were carried out on two sets of real-world HSIs, and it was found that MSPSW-CEM generally showed a better detection performance than other object detection methods.

Residential land (RL), as a typical kind of urban functional zone, plays an important role in urban planning and land census. Recent years have witnessed frequent changes in RL via the process of urbanization. The extraction of RL from high spatial resolution optical images can reflect the status quo of land use/land cover to a certain extent, which is of great significance to land census and urban planning. We adopt a scene classification strategy to extract RL and mainly focus on the extraction of four common types of RL in China: old-style village, low-density high-rise, medium-density low-rise, and low-density low-rise. We design a multifeature hierarchical (MFH) algorithm for RL extraction. First, RL is extracted based on the gray level concurrence matrix and a fuzzy classification algorithm. Then an improved bag-of-visual-words algorithm is introduced to further realize the extraction of RL. The effectiveness of our proposed method is analyzed with a sample dataset and large images. We also analyze the separability among different kinds of RL. We compare the experimental results with those of three other algorithms, and the results demonstrate that the MFH algorithm performs better in terms of the accuracy and efficiency of the RL extraction. The results can provide services for land surveying and urban planning, and the technological processes and experimental design in the algorithm can provide a reference for the research in related fields.

Due to the spectrum of a moving target being almost submerged in ground clutter, it is difficult for high-speed platform synthetic aperture radar to indicate a slowly moving target by traditional methods. To solve this problem, a detection and azimuth velocity estimation method is proposed based on along-track dual-beam synthetic aperture radar mode. Back projection is adopted to focus the forward- and backward-looking-beam images due to abilities such as automatic coregistration and geometric correction. The positions of the moving target in the two images are deduced, and we show that there is an azimuth location offset for the moving target in the forward-and-backward-looking images proportional to its azimuth velocity, which can be used to detect the moving target and estimate its azimuth velocity. Furthermore, a refocusing method is proposed to obtain a more accurate estimation. Simulation results show that the proposed method can detect slowly moving targets, especially those only with an azimuth velocity, which is difficult for many existing methods. In addition, the accuracy of azimuth velocity estimation can reach 0.01  m  /  s when the signal-to-clutter ratio is greater than 5 dB under simulation conditions.

Ground penetrating radar (GPR) is valuable for the detection of subsurface objects with little or no metal content, such as plastics, ceramics, and concrete piping. However, the effects of antenna configuration parameters, such as height and angle, are not well studied for all sensing applications. GPR simulations and laboratory GPR experiments are performed to evaluate the effects of antenna angle and height on the sensitivity of bistatic air-launched GPR, to search for buried nonmetallic objects. The results presented provide guidance for the development of air-launched GPR systems installed on unmanned aerial vehicles for in-flight subsurface scanning of buried targets.

The lunar surface consists of rocks of varying sizes and shapes, which are made of minerals, such as pyroxene, plagioclase, olivine, and ilmenite, that exhibit distinctive spectral characteristics in the visible and near-infrared (VIS–NIR) and short-wave infrared (SWIR) regions. To analyze the composition of the lunar surface minerals, several spectrometers based on acousto-optic tunable filters (AOTFs) have been developed to detect lunar surface objects and to obtain their reflectance spectra and geometric images. These spectrometers, including the VIS–NIR imaging spectrometer onboard China’s Chang’e 3/4 unmanned lunar rovers and the Lunar Mineralogical Spectrometer onboard the Chang’e 5/6 lunar landers, use AOTFs as dispersive components. Both are equipped with a VIS/NIR imaging spectrometer, one or several SWIR spectrometers, and a calibration unit with dust-proofing functionality. They are capable of synchronously acquiring the full spectra of the lunar surface objects and performing in-situ calibrations. We introduce these instruments and present a brief description of their working principle, implementation, operation, and major specifications, in addition to the initial scientific achievement of lunar surface exploration.

With the rapid development of three-dimensional (3-D) laser scanners, the correspondent point cloud accuracy and density are continuously improving. However, the point data volume becomes larger and larger, which brings new challenges for the point cloud data real-time processing on personal computers. To meet the managing requirement for real-time point cloud processing, we proposed a hybrid index model characterized by top-down greedy splitting (TGS) R-tree and 3-D quadtree, aiming at the balance improvement and the high index query efficiency. First, the large-scale point cloud data are divided into grids based on their spatial distribution, and then, the proposed TGS R-tree algorithm is applied to organize these grids. Second, a 3-D quadtree local index model is developed to manage local points in each grid. Experiments of five point cloud data scenarios are conducted to evaluate the proposed method, and the results show that the proposed model can meet the efficiency need of various 3-D point cloud managements, especially for those mainly distributed in XY plane, including the airborne LiDAR point cloud.