ALBERT

Description: The capabilities of bistatic radar observations to estimate the wind field over the ocean are investigated in this paper. The work is based on the analysis of simulated data obtained through a well-established electromagnetic model, which accounts for the anisotropy of the ocean’s spectrum and of second-order effects of the scattering phenomenon. Both co-polarized and cross-polarized C-band numerical data, obtained considering monostatic and bistatic configurations, are exploited to investigate on the existence of optimal configurations able to minimize the wind vector error estimation. To this aim, the sensitivities of the bistatic normalized radar cross section with respect to both wind speed and direction are accurately investigated and exploited to evaluate the minimum achievable error standard deviation of the estimation. Small and large baselines are analyzed, giving particular emphasis to bistatic geometries constituted by one or two passive receivers aligned along the track defined by the active system. This investigation, originally performed in the framework of the SAOCOM-CS scientific satellite mission, is conceived to accurately assess the potentiality of bistatic observations of the ocean over variable baselines and to gather valuable information for the design of future bistatic satellite missions.

Description: The normalized differential spectral attenuation (NDSA) approach was proposed years ago as an effective way to estimate the integrated water vapor (IWV) along a tropospheric propagation path between two low Earth orbit satellites. Two applications are possible: the retrieval of vertical profiles of WV if the sense of rotation is opposite and the retrieval of 2-D fields of WV over vertical tropospheric sections if the sense is the same. The method relies on the measurement of the so-called spectral sensitivity $S$ at given frequencies, and on IWV-S relationships that convert $S$ into an estimate of IWV along the radio link where $S$ is measured. In this paper, we recompute the IWV-S relationships using synthetic atmospheres generated by means of European Centre for Medium-Range Weather Forecasts (ECMWF) analysis data instead of radiosonde profiles as done by ourselves in the past. Thanks to the uniform spatial distribution of the ECMWF data on a global Earth scale, we were able to validate the IWV-S relationships in the Ku/K band previously found through synthetic atmospheres generated by means of the aforementioned irregularly spaced radiosonde data, and to define the IWV-S relationships at 179 and 181 GHz that are exploitable in the upper troposphere. Since the ECMWF data also include information about the liquid water (LW) content, we then show that an additional $S$ channel at 32 GHz can be exploited to detect and correct the bias induced by LW on IWV estimates made by applying the NDSA in the Ku/K band.

Description: Presents the cover/table of contents for this issue of the periodical.

Description: Provides a listing of current staff, committee members and society officers.

Description: The seasonal impact of adjacency effects (AE) on satellite ocean color data at visible and near-infrared (NIR) wavelengths by the Sea-Viewing Wide Field-of-View Sensor, the Moderate Resolution Imaging Spectroradiometer onboard the Aqua platform (MODISA), the Medium Resolution Imaging Spectrometer, the Ocean and Land Color Instrument, the Operational Land Imager (OLI), and the MultiSpectral Imagery (MSI) was theoretically evaluated at a validation site in the northern Adriatic Sea. The analysis made use of comprehensive simulations accounting for multiple scattering, sea surface roughness, sensor viewing geometry, actual coastline, typical and extreme atmospheric conditions, and the seasonal variability of solar illumination and, land and water optical properties. Results, obtained by relying on the normalization of the radiometric sensitivity of each sensor to the same input radiance, show that the spectral and seasonal impacts of AE considerably vary among sensors. AE significantly exceed the radiometric sensitivity of MSI at its sole blue band in winter, whereas they significantly outdo the noise threshold of OLI and MODISA high-resolution data exclusively in the NIR in summer. Conversely, for all other sensors and for MODISA low-resolution data, AE are particularly significant at NIR bands between March and October and at the blue–green bands in winter.

Description: In this letter, a localization algorithm, which combines energy operator with Doppler processing, is proposed for Doppler radar human sensing applications. For this algorithm, the energy operator is first used to extract the target components of interest from radar echoes and estimate their instantaneous frequencies (IFs). Then, on the basis of the IF estimation result, Doppler processing is applied to synthesize the target movement trajectories. Compared with the traditional localization methods, the proposed algorithm can more precisely estimate the target movement trajectory. Besides, it can further avoid the frequency ambiguity issue, and thus can be very promising for multitarget sensing applications. Experimental results are shown to demonstrate the performance of the proposed algorithm.

Description: In this letter, we introduce an asymmetric adaptation neural network (AANN) method for cross-domain classification in remote sensing images. Before the adaptation process, we feed the features obtained from a pretrained convolutional neural network to a denoising autoencoder (DAE) to perform dimensionality reduction. Then the first hidden layer of AANN (placed on the top of DAE) maps the labeled source data to the target space, while the subsequent layers control the separation between the available land-cover classes. To learn its weights, the network minimizes an objective function composed of two losses related to the distance between the source and target data distributions and class separation. The results of experiments conducted on six scenarios built from three benchmark scene remote sensing data sets (i.e., Merced, KSA, and AID data sets) are reported and discussed.

Description: This letter shows the phenomena of Brewster angle damping and its implication observed in the synthetic aperture radar (SAR) images of concrete constructions, such as a bridge and seawalls over the sea. The Fresnel reflection coefficient of concrete material is close to zero at the Brewster angle for X-band V-polarization microwave. The TerraSAR-X images of Tokyo Bay, Japan, at small incidence angles (20.1°–21.4°) showed strong double-bounce reflection between the sea surface and coastal structure with HH-polarization, whereas very little radar backscatter was observed with VV-polarization. The same little radar backscatter was seen in the images of concrete walls on ground and swamp areas covered with reeds. This effect is illustrated with HH/VV intensity and phase difference images, and ground survey data; its implication is also suggested for a better understanding of polarimetric SAR images.

Description: Accurate and fast infrared (IR) foreground object detection is one of the most significant issues to be solved due to its important meaning for IR target recognition, IR precise guidance, IR video surveillance, and so on. A common approach for such tasks is “background subtraction,” which aims to detect foreground object through background modeling. Thus far, many background subtraction methods have been proposed and have achieved good performance. However, due to the special characteristics of IR images, a few algorithms are suitable for IR foreground object detection. Recently, features learned from convolutional neural networks (CNNs) have demonstrated great success in many vision tasks, such as classification and recognition. In this letter, we propose a novel multiscale fully convolutional network architecture for IR foreground object detection. Given a CNN model pretrained on a large-scale image data set, our method takes output features from different layers of the network. With features from multiple scales, our feature representation contains both category-level semantics and fine-grain details. The experimental results on IR image sequences show that the proposed method achieves the state-of-the-art performance while operating in real time.

Description: Provides a listing of current staff, committee members and society officers.

Description: Presents the table of contents for this issue of this publication.

Description: The reconstruction of urban buildings from large-scale airborne laser scanning point clouds is an important research topic in the geoscience field. Large-scale urban scenes usually contain a large number of object categories and many overlapped or closely neighboring objects, which poses great challenges for classifying and modeling buildings from these data sets. In this paper, we propose a deep reinforcement learning framework that integrates a 3-D convolutional neural network, a deep Q-network, and a residual recurrent neural network for the efficient semantic parsing of large-scale 3-D point clouds. The proposed framework provides an end-to-end automatic processing method that maps the raw point cloud to the classification results of the given categories. After obtaining the building classes, we utilize an edge-aware resampling algorithm to consolidate the point set with noise-free normals and clean preservation of sharp features. Finally, 2.5-D dual contouring, which is a data-driven approach, is introduced to generate urban building models from the consolidated point clouds. Our method can generate lightweight building models with arbitrarily shaped roofs while preserving the verticality of connecting walls.

Description: This paper shows initial results from estimating Doppler radial surface velocities (RVLs) over Arctic sea ice using the Sentinel-1A (S1A) satellite. Our study presents the first quantitative comparison between ice drift derived from the Doppler shifts and drift derived using time-series methods over comparable time scales. We compare the Doppler-derived ice velocities with global positioning system tracks from a drifting ice station as well as vector fields derived using traditional cross correlation between a pair of S1A and Radarsat-2 images with a time lag of only 25 min. A strategy is provided for precise calibration of the Doppler values in the context of the S1A level-2 ocean RVL product. When comparing the two methods, root-mean-squared errors (RMSEs) of 7 cm/s were found for the extra wide (EW4) and EW5 swaths, while the highest RMSE of 32 cm/s was obtained for the EW1 swath. Though the agreement is not perfect, our experiment demonstrates that the Doppler technique is capable of measuring a signal from the ice if the ice is fast moving. However, for typical ice speeds, the uncertainties quickly grow beyond the speeds we are trying to measure. Finally, we show how the application of an antenna pattern correction reduces a bias in the estimated Doppler offsets.

Description: The bistatic radar equation currently used for simulating surface-reflected waveforms or delay-Doppler maps (DDMs), produced by signals of opportunity from global navigation satellites system (GNSS) or communication satellites, was previously derived under some limiting assumptions. One of them was the use of the Kirchhoff approximation in a geometric optics limit that assumes strong diffuse (noncoherent) scattering typical for very rough surfaces. This equation would produce an incorrect result for the case of weak diffuse scattering, or in the presence of coherent reflection. In this paper, it is shown that the assumption of strong diffuse scattering is not necessary in deriving such an equation. The derivation of a generalized bistatic radar equation is now based only on the assumption of roughness statistics being spatially homogeneous, and thus this equation is applicable for a much wider range of surface conditions and scattering geometries. This approach allows to correctly describe the transition from partially coherent scattering to completely noncoherent, strong diffuse scattering. It is demonstrated for the case of the GNSS-R DDMs simulated for a wide range of surface winds, and their transitional behavior is discussed.

Description: Differential interferometric synthetic aperture radar (InSAR) time-series processing relies on identifying coherent pixels in SAR image stacks that show the persistent scatterer (PS) or distributed scatterer (DS) behavior. Accuracy of InSAR time-series estimates is dependent on the quality of selected PS/DS pixels. Current pixel selection techniques perform well when identifying highly coherent pixels but produce many false alarms in low coherence regions due to the inherent bias in residual phase estimation. Therefore, pixels with low coherence may have the appearance of noise and be rejected if the coherence threshold is too high. In contrast, lowering the threshold increases the number of false alarms introduced in processing giving noisier time-series as a result of incorrect phase unwrapping. The multidimensional SAR data acquisition can be described as a zero mean Gaussian process fully described by the covariance matrix. In this paper, we investigate the covariance matrix using a random matrix theory approach to find the statistical properties of the eigenvalues for simulated and real SAR data. The probability distribution of all the eigenvalues in this case is limited by the Marcenko–Pastur distribution. The histogram of the highest eigenvalue follows a Tracy–Widom distribution. Thus, by adopting a pixel selection strategy based on a threshold on the highest eigenvalue of the coherence matrix, we can differentiate between low coherence and noise pixels. In addition, our technique provides a methodology to detect the number of targets present in multiscatterer layover pixels and extract time-series information from double bounce response of bridges. Applying the technique for TerraSAR-X data over Berlin shows the effectiveness of the algorithm.

Description: Logging-while-drilling (LWD) or measuring-while-drilling tools are routinely used to guide well placement during exploration of hydrocarbons reservoirs. These tools have become fundamental for directional and horizontal drilling operations. In this paper, we present a perturbation method to model electromagnetic fields produced by time-harmonic sources in radially stratified and axially toroidal structures describing LWD tools inside curved boreholes (i.e., boreholes with axial bending). The proposed formulation is validated against brute-force finite-difference results in various representative scenarios. Numerical results indicate that the proposed perturbative approach can provide a reduction in the computational effort required to analyze this class of problems of several orders of magnitude versus brute-force approaches.

Description: Due to the 3-D nature of hyperspectral images, as well as the spatial properties (such as regularity and continuity) of land covers, many 3-D feature extraction operators have been designed to fully exploit the joint spatial–spectral information. However, the large amount of obtained features can suffer from the “curse of dimensionality” problem, especially for the small training sample set. Moreover, various spatial–spectral features can represent the characteristics of the hyperspectral image from different aspects. In this paper, a multiple 3-D feature fusion framework (M3DF 3 ) has been proposed for hyperspectral image classification. First, we extend the 2-D Gabor surface feature into 3-D (3DSF) domains to comply with the spatial–spectral structure of the hyperspectral image, which is directly applied on the original hyperspectral image instead of the Gabor features. Second, three 3-D feature extraction methods, including the 3-D morphological profile, the 3-D local binary pattern, and the proposed 3DSF, that, respectively, characterize the hyperspectral image from three different angles, i.e., morphology, local dependence, and shape smoothness, are fused under a multitask sparse representation framework to take full advantage of the multiple 3-D features together. The proposed M3DF 3 approach was fully tested on three real-world hyperspectral image data, i.e., the widely used Indian Pines, Pavia University, and Houston University. The results show that our method can achieve as high as 68.22%, 79.44%, and 72.84% accuracies, respectively, even when only few samples, i.e., three samples per class, are used for training.

Description: Presents the front cover for this issue of the publication.

Description: The conventional methods for target detection and discrimination in high-resolution synthetic aperture radar (SAR) images usually have low accuracy and slow speed, especially for large complex scenes. To overcome these drawbacks, in this paper, we propose a target detection and discrimination method based on visual attention model. In the detection stage, to pop out the targets and suppress the background clutter in the saliency map, we select the task-dependent scales from the Gaussian pyramid of the original SAR image. Moreover, we adopt the clustering algorithm to remerge several isolated focus of attention areas, which are obtained from the saliency map, into a complete target region. The candidate target SAR image chips are extracted with relative high accuracy and low time cost in this stage. Since there may be single target, multiple targets, or partial targets with complex clutter in each SAR image chip, it is hard to acquire accurate target-shaped blob via segmentation. Some classical discrimination features which are extracted based on target segmentation may lose effectiveness. In the discrimination stage of our method, to solve the above problem, based on the saliency and gist (SG) features for optical satellite images, we propose the modified SG (MSG) features for SAR target discrimination. The MSG features are complementary to each other and can provide a more complete description of the extracted SAR image chips without segmentation, which also reduces the computation burden. The experimental results on the synthetic images and miniSAR real SAR image data set demonstrate that the proposed target detection and discrimination method can detect and discriminate the targets from the complex background clutter with high accuracy and fast speed in high-resolution SAR images.

Description: Hyperspectral image (HSI) noise reduction is an active research topic in HSI processing due to its significance in improving the performance for object detection and classification. In this paper, we propose a joint spectral and spatial low-rank (LR) regularized method for HSI denoising, based on the assumption that the free-noise component in an observed signal can exist in latent low-dimensional structure while the noise component does not have this property. The proposed HSI denoising method not only considers the traditional LR property across the spectral domain but also leverages nonlocal LR property over the spatial domain. The main contribution of this paper is the incorporation of the low-rankness-based nonlocal similarity into sparse representation to characterize the spatial structure. Specially, the similar patches in each cluster usually contain similar sharp structure such as edges and textures; LR performed on cluster entitles to achieve a lower rank than that on the global spectral correlation. To make the proposed method more tractable and robust, we develop a variable splitting-based technique to solve the optimization problem. Experiment results on both simulated and real hyperspectral data sets demonstrate that the proposed method outperforms state-of-the-art methods with significant improvements both visually and quantitatively.

Description: Downward-looking linear array 3-D synthetic aperture radar (SAR) has attracted increasing attention in the field of radar imaging. As widely reported, the volume of data can be significantly reduced by a random sparse linear array. However, the 2-D under-sampled azimuth-cross-track data brought by the sparse linear array will produce high-level side-lobes, as well as the aliasing and the false-alarm targets. To deal with those problems, this paper introduces a recently developed theory, matrix completion (MC). The new theory could recover a matrix with a small subset of known elements of the matrix. It is founded on the assumption that the matrix is essentially low rank. For downward-looking 3-D SAR with a random sparse linear array, the received 3-D data can be treated as a series of uncorrelated 2-D matrices by the separated channel process. First, range compression can be realized by means of pulse compression. Then, the sets of the 2-D under-sampled azimuth-cross-track matrix can be completed into a full-sampled one via MC trick. The resulting 3-D images can be focused by synthetic aperture technique along the azimuth direction and beamforming operation along the cross-track direction, with the recovered full-sampled matrix. The proposed algorithm achieves high resolution and low-level side-lobes with the acceptable computational cost and memory consumption. It is verified by several numerical simulations and multiple comparative studies on real data. The experimental results clearly demonstrate the imaging performance across different under-sampling rates and signal-noise rates.

Description: Hydraulic fracturing is a technique to fracture rocks by pumping high-pressure fluid into a segment of a well. The created fractures help to release a hydrocarbon resource such as oil or natural gas from the rock. A group of small-scaled fracturing field tests are performed by the Advanced Energy Consortium to investigate the feasibility of using the galvanic electromagnetic (EM) method to map fractures. The injected proppants are designed with high EM contrasts (e.g., conductivity and permittivity) to generate detectable signals at electrode-type sensors. To map the created fractures, an efficient 3-D EM inversion method is introduced to simultaneously reconstruct conductivity and permittivity profiles in fractures. First, to test the capability of the inversion solver and the designed experimental setting for successful fracture mapping, the noise-polluted synthetic data are used to reconstruct the fracture on a theoretical model. It shows that the designed experimental setting can be used to map the fracture and the inversion solver is able to reconstruct the fracture in both conductivity and permittivity. The inversion method is then applied to two hydraulic fracturing field tests with injected high-contrast proppants, Loresco coke breeze and steel shot. The fracture conductivity and permittivity are reconstructed based on the voltage signals difference between the postfracturing and prefracturing data. The reconstructed fracture profiles are compared with the coring samplings to show the reliability of the inversion results. Their good agreement demonstrates that the experimental setting and the galvanic inverse solver are able to estimate the fracture size and location reliably.

Description: One option for the improvement of weather radar technology is the use of dual-polarized phased-array radar for weather observations. Several risk factors on this path have been identified and one of the most important ones is the existence of significant cross-polar patterns inherent to the phased-array antenna. These antenna patterns induce cross-coupling between returns from the two orthogonal radiation planes, which results in the biases of polarimetric variable estimates. Furthermore, the inductive and capacitive coupling in hardware behind the antenna may exacerbate the cross-coupling effects. This presents a formidable challenge because sufficient cross-polar isolation is difficult to achieve by the antenna hardware alone. Hence, additional approaches are required to reduce the biases due to cross-coupling. One proposed technique is a 180° pulse-to-pulse phase change of signals injected in either the horizontal or vertical ports of the transmission elements. This approach was analyzed for signals processed in the time domain but its effects in the frequency domain have not been investigated. Herein, these effects are considered in the presence of nondepolarizing scatterers.

Description: Leaf area index (LAI) and fraction of photosynthetically active radiation (FPAR) absorbed by vegetation have been successfully generated from the Moderate Resolution Imaging Spectroradiometer (MODIS) data since early 2000. As the Visible Infrared Imaging Radiometer Suite (VIIRS) instrument onboard, the Suomi National Polar-orbiting Partnership (SNPP) has inherited the scientific role of MODIS, and the development of a continuous, consistent, and well-characterized VIIRS LAI/FPAR data set is critical to continue the MODIS time series. In this paper, we build the radiative transfer-based VIIRS-specific lookup tables by achieving minimal difference with the MODIS data set and maximal spatial coverage of retrievals from the main algorithm. The theory of spectral invariants provides the configurable physical parameters, i.e., single scattering albedos (SSAs) that are optimized for VIIRS-specific characteristics. The effort finds a set of smaller red-band SSA and larger near-infrared-band SSA for VIIRS compared with the MODIS heritage. The VIIRS LAI/FPAR is evaluated through comparisons with one year of MODIS product in terms of both spatial and temporal patterns. Further validation efforts are still necessary to ensure the product quality. Current results, however, imbue confidence in the VIIRS data set and suggest that the efforts described here meet the goal of achieving the operationally consistent multisensor LAI/FPAR data sets. Moreover, the strategies of parametric adjustment and LAI/FPAR evaluation applied to SNPP-VIIRS can also be employed to the subsequent Joint Polar Satellite System VIIRS or other instruments.

Description: Ultrawideband radar with high-range resolution is a promising technology for use in short-range 3-D imaging applications, in which optical cameras are not applicable. One of the most efficient 3-D imaging methods is the range-point migration (RPM) method, which has a definite advantage for the synthetic aperture radar approach in terms of computational burden, high accuracy, and high spatial resolution. However, if an insufficient aperture size or angle is provided, these kinds of methods cannot reconstruct the whole target structure due to the absence of reflection signals from large part of target surface. To expand the 3-D image obtained by RPM, this paper proposes an image expansion method by incorporating the RPM feature and fully polarimetric data-based machine learning approach. Following ellipsoid-based scattering analysis and learning with a neural network, this method expresses the target image as an aggregation of parts of ellipsoids, which significantly expands the original image by the RPM method without sacrificing the reconstruction accuracy. The results of numerical simulation based on 3-D finite-difference time-domain analysis verify the effectiveness of our proposed method, in terms of image-expansion criteria.

Description: This paper addresses the performance in the retrieval of 3-D mean deformation maps by exploiting simultaneous or quasi-simultaneous squinted synthetic aperture radar (SAR) interferometric acquisitions in a repeat-pass scenario. In multisatellite or multibeam low earth observation missions, the availability of two (or more) lines of sight (LOSs) allows the simultaneous acquisition of SAR images with different squint angles, hence improving the sensitivity to the north–south component of the deformation. Due to the simultaneity of the acquisitions, the troposphere will be highly correlated and, therefore, will tend to cancel out when performing the differential measurement between the interferograms obtained with the different LOSs, hence resulting in a practically troposphere-free estimation of the along-track deformation measurement. In practice, however, the atmospheric noise in the differential measurement will increase for increasing angular separations. This paper expounds the mathematical framework to derive the performance by properly considering the correlation of the atmospheric delays between the simultaneous acquisitions. To that aim, the hybrid Cramér–Rao bound is exploited making use of the autocorrelation function of the troposphere. Some performance examples are presented in the frame of future spaceborne SAR missions at C and L band.

Description: The detection of meteorological targets using ground-based weather radars usually suffers from ground clutter and beam blockage. These nonmeteorological or weakened signals should be identified so quality control should be implemented before weather radar data can be used. Conventional quality control methods aim at differentiating between echo structures of ground clutter and meteorological targets, and use terrain information to calculate beam blockage regions based upon standard atmospheric refraction. However, it is difficult to achieve the goal for long-term large data sets by conventional methods due to the complexity and diversity of weather radar echoes. In this paper, regions of ground clutter and beam blockage are first identified through the statistics on spatial distribution of reflectivity and fuzzy logic classification, and then they are used as masks to remove data from the scan. The new method is applied to data of the Nanjing weather radar in China. By the aid of a proposed evaluation scheme and the visual recognition, quality control results of the new method are compared with those of the conventional methods. It is found that the new method can provide better identification of ground clutter or beam blockage and thus better quality control results. The new scheme has a good prospect in operational service for its principle advantages, easy applicable conditions, and better performance compared with conventional methods.

Description: This paper describes the adaptation of a Bayesian sea ice detection algorithm for the scatterometer on-board the European Remote Sensing (ERS) satellites (ERS-1 and ERS-2). The algorithm is based on statistics of distances to ocean wind and sea ice geophysical model functions (GMFs) and its performance is validated against coincident active and passive microwave data. We furthermore propose a new model for sea ice backscatter at the C-band in vertical polarization based on the sea ice GMFs derived from ERS and advanced scatterometer data. The model characterizes the dependence of sea ice backscatter on the incidence angle and the sea ice type, allowing a more precise incidence angle correction than afforded by the usual linear transformation. The resulting agreement between the ERS, QuikSCAT, and special sensor microwave imager sea ice extents during the year 2000 is high during the fall and winter seasons, with an estimated ice edge accuracy of about 20 km, but shows persistent biases between scatterometer and radiometer extents during the melting period, with scatterometers being more sensitive to summer (lower concentration and rotten) sea ice types.

Description: This paper evaluates the calibration quality during the blackbody (BB) warm-up cool-down cycle for thermal emissive bands onboard Aqua Moderate Resolution Imaging Spectroradiometer (MODIS) and Suomi National Polar-orbiting Partnership (S-NPP) Visible Infrared Imaging Radiometer Suite (VIIRS). This evaluation utilizes data from Aqua MODIS Collection 6 Level-1B products and VIIRS Sensor Data Records in 6-min granule format provided by the NASA Land Science Investigator-led Processing System. Nearly simultaneous hyperspectral measurements from the Aqua Atmospheric Infrared Sounder (AIRS) and the S-NPP Cross-track Infrared Sounder (CrIS) are used as references for MODIS and VIIRS, respectively. Each AIRS footprint of 13.5 km is co-located with multiple MODIS pixels while each CrIS field of view of 14 km is co-located with multiple VIIRS pixels. The corresponding AIRS-simulated MODIS and CrIS-simulated VIIRS radiances are derived by convolutions based on sensor-dependent relative spectral response functions. In this paper, the analysis mainly focuses on the bands that are used in sea surface temperature products. The results show that there is virtually no impact for MODIS bands 22 and 23 and bands 31 and 32 for a BB temperature below 290 K; however, when the BB temperature increases above 290 K, the impact is up to 0.3 K for bands 22 and 23 and 0.05 K for bands 31 and 32, respectively. For VIIRS, BB temperature-dependent drifts are observed in M15 and M16, which can reach 0.15 and 0.1 K, respectively, over the operational BB temperature range and the VIIRS brightness temperature range.

Description: Most of the existing deep-learning-based methods are difficult to effectively deal with the challenges faced for geospatial object detection such as rotation variations and appearance ambiguity. To address these problems, this paper proposes a novel deep-learning-based object detection framework including region proposal network (RPN) and local-contextual feature fusion network designed for remote sensing images. Specifically, the RPN includes additional multiangle anchors besides the conventional multiscale and multiaspect-ratio ones, and thus can deal with the multiangle and multiscale characteristics of geospatial objects. To address the appearance ambiguity problem, we propose a double-channel feature fusion network that can learn local and contextual properties along two independent pathways. The two kinds of features are later combined in the final layers of processing in order to form a powerful joint representation. Comprehensive evaluations on a publicly available ten-class object detection data set demonstrate the effectiveness of the proposed method.

Description: Recently, multifeature learning in collaborative representation classification (CRC) for hyperspectral images has generated promising performance. In this paper, two novel multifeature learning algorithms that update dictionary directly and indirectly are proposed. In order to offer the complementarity of multifeature, four different types of features—global feature (i.e., Gabor feature), local feature (i.e., local binary pattern), shape feature (i.e., extended multiattribute profiles), and spectral feature—are adopted in this paper. Under the hypothesis that most of the features should share the same coding pattern in CRC, this paper proposes to learn proper dictionaries for each feature until obtaining stable codes in a linear classifier. Furthermore, to avoid the explicit mapping of infinite-dimensional dictionaries in a nonlinear kernelized classifier, an indirect approach to construct the transformation matrix from original dictionaries to learn new dictionaries is developed. Three real hyperspectral images acquired from different sensors are adopted for performance evaluation. The experimental results demonstrate that the proposed methods can provide superior performance compared with those of the state-of-the-art classifiers.

Description: Advertisement, IEEE.

Description: Ground penetrating radar imaging from the data acquired with arbitrarily oriented dipole-like antennas is considered. To take into account variations of antenna orientations resulting in spatial rotation of antenna radiation patterns and polarizations of transmitted fields, the full-wave method that accounts for the near-, intermediate-, and far-field contributions to the radiation patterns is applied for image reconstruction, which is formulated as a linear inversion problem. Two approaches, namely, an interpolation-based method and a nonuniform fast Fourier transform-based method, are suggested to efficiently implement the full-wave method by computing exact Green’s functions. The effectiveness and accuracy of the method proposed have been verified via both numerical simulations and experimental measurements, and significant improvement of the reconstructed image quality compared with the traditional scalar-wave-based migration algorithms is demonstrated. The results can be directly utilized by forward-looking microwave imaging sensors such as installed at tunnel boring machine or can be used for the observation matrix computation in regularization-based inversion algorithms.

Description: Advertisement, IEEE.

Description: Advertisements.

Description: This paper presents an extension of H/A/Alpha polarimetric SAR (PolSAR) decomposition for polarimetric interferometric SAR (PolInSAR) images, and introduces new parameters using both polarimetric and interferometric data. These parameters provide more information than extracted parameters from H/A/Alpha PolSAR decomposition. The new parameters significantly increase the potential of PolSAR data for, for example, forest and oriented building discrimination. With the availability of interferometric information in addition to PolSAR information, there is the possibility to reduce the entropy of PolSAR data. The relationship between the entropy of PolInSAR decomposition and coherence between images has also been shown. E-SAR PolInSAR L-band data of Oberpfaffenhofen, Germany, Northern Sweden, and Goermin, Germany are used to validate the H/A/Alpha PolInSAR decomposition. The experimental results on the first data set, including agricultural, forest, and urban areas, show that this decomposition has a better performance than the standard H/A/Alpha decomposition method in oriented urban areas with large orientation angle, for example. Also, the results on the second data set of a forest area show that the PolInSAR decomposition has more reasonable and superior performance over the PolSAR decomposition. The forest and nonforest regions can be correctly discriminated by using the presented PolInSAR parameters. Finally, in the last data set, the better detection of the agricultural crops and fields boundaries is provided by the proposed method.

Description: An investigation of the eddy and coastal Kelvin wave activities in the Bay of Bengal (BoB) is carried out during premonsoon season in two years of Indian summer monsoon deficit in June (2009 and 2012), occurred in the recent warming hiatus period. Using altimeter observations, our study reveals that over the northern BoB cyclonic eddy kinetic energy is reduced by 35% and 50% from the climatology during premonsoon seasons in 2009 and 2012, respectively, while the cyclonic eddy area is reduced by 18% and 24%, respectively. A concurrent reduction is observed in the first upwelling Kelvin wave (uKW) activities in the eastern equatorial Indian Ocean as well as in the coastal BoB for these years. The reduction in the generation of the first uKW in the eastern equatorial Indian Ocean is attributed to the westerly wind anomalies in January–March of these years. Additionally, meridional wind stress anomalies during March–April in these years are found to be southerly, causing anomalous coastal downwelling in the eastern rim of BoB. This coastal downwelling blocks the propagation of the first uKW. The decrease in the first uKW activities in the coastal waveguide of the BoB reduces the radiation of upwelling Rossby waves, thereby decreasing the cyclonic eddy activities in the northern BoB. The results from this letter could be helpful for further understanding of upper ocean mixing processes in the BoB during monsoon deficit years.

Description: A bi-iterative physical optics (PO) method is presented in this letter to investigate the composite scattering from a perfectly electric conducting object partially buried in the dielectric rough sea surface. In the present method, both the scatterings of a partially buried object and the underlying sea surface are calculated by the PO method. And a bi-iterative strategy is considered, including the mutual interaction among points on the object in direct scattering and the mutual interaction between the object and the sea surface in coupling scattering. In addition, the coupling interaction between the partially buried object and the sea surface contains two parts: 1) the upper surface of the sea surface and the upper part of the object and 2) the lower surface of sea surface and the lower part of the object. In numerical simulations, the bistatic normalized radar cross sections of the composite model are computed by the bi-iterative PO method and are compared with those by the conventional method of moments for different object types and polarizations. The results show that the proposed method has a good accuracy and can greatly reduce the computational time and memory requirement.

Description: A novel tensorization framework is proposed, which utilizes the Kronecker product to combine multifrequency polarimetric synthetic aperture radar data in conjunction with an artificial neural network (ANN) for classification. The ANN comprises of two stages, where an unsupervised stochastic sampling autoencoder learns an efficient representation and a supervised feed forward network performs classification. The proposed framework is demonstrated using multifrequency (C-, L-, and P-bands) data sets collected by the AIRSAR system. The classification performance of single tensor product of dual- and triple-band combinations is evaluated. It is observed that the classification accuracy of the tensor products outperforms single, as well as, the simple augmentation of the frequency bands.

Description: The combination of linear range walk correction and keystone transform is a good choice to focus high-resolution highly squint synthetic aperture radar (SAR) data because it is an effective way to remove linear range cell migration (RCM) completely and mitigate range–azimuth coupling. However, the results of this kind of imaging algorithm produce 2-D-variant residual RCM and variant-dependence Doppler phases. To obtain high-quality SAR image, an improved imaging algorithm using an azimuth-variant residual RCM correction (RCMC) and an extended nonlinear chirp scaling (ENLCS) is proposed in this letter. A new circle model is constructed to analyze the azimuth-variant properties of the residual high-order RCM and the Doppler phases. Based on this circle model, an azimuth-variant residual RCMC is implemented by multiplying a fourth-order phase function, and an improved ENLCS is derived to accomplish the azimuth equalization for azimuth compression. Simulation results validate the excellent performance of the proposed algorithm.

Description: In this letter, we introduce adaptive probability thresholding in addition to our previously developed technique for automated detection of ice and open water from RADARSAT-2 ScanSAR dual-polarization HH–HV images. Situations where the probability threshold needs to be modified were identified based on the analysis of misclassified ice and water samples when the static probability threshold of 0.95 is applied. We found that with the use of the proposed approach, the fraction of misclassified ice samples decreased from 0.98% to 0.24% and the fraction of misclassified water samples decreased from 0.35% to 0.09% in the most clean verification scenario against Canadian Ice Service Image Analysis pure ice and water data, while the fraction of correctly classified ice and water samples did not decrease appreciably, from 72.2% to 65.4%. The developed approach will be implemented as a part of the data assimilation component of the operational Environment and Climate Change Canada Regional Ice-Ocean Prediction System.

Description: Recent studies have revealed that the residual range migration effects of synthetic aperture radar (SAR) imagery smears induced by moving targets can exhibit complicated shapes that are not limited to that of parabolas. This letter demonstrates that automatic focusing methods can remove such range migration effects by estimating and compensating for phase errors directly in the radar video phase history domain. This approach is validated using measured Ku-band SAR clutter data containing buildings and foliage.

Description: In this letter, we propose a superpixel-level target detection approach for synthetic aperture radar (SAR) images. With superpixel segmentation, SAR image is divided into meaningful patches and more statistical information can be provided in superpixels compared with single pixels. The statistical difference between target and clutter superpixels can be measured with the intensity distributions of pixels in them. With the assumption of SAR data obeying Gamma distribution, the superpixel dissimilarity is defined. With this basis, the global and local contrast can be obtained and integrated to enhance target and suppress clutter simultaneously. Thus, better target detection performance can be achieved. Different from traditional target detection schemes based on backscattering difference between target and clutter pixels, the proposed method relies on the statistical difference of superpixels. The effectiveness of the proposed method can be demonstrated with experimental results on real SAR images.

Description: This letter presents an analysis of the temporal characteristics of electromagnetic waves scattered from a time-varying reservoir surface at low grazing angles. The data collection campaigns were conducted using a polarimetric S-band radar at Wachusett Reservoir in MA, USA, and VV and HH polarized radar returns were simultaneously captured. The temporal behavior of the backscattering from the reservoir surface was analyzed for 180 distinct radar geometries, focusing in particular on the impact of polarization, radar geometry, and wind condition. To understand the shape of the Doppler spectrum, the power spectral density is estimated by a periodogram. In addition, decorrelation time, Doppler centroid, and variance are estimated and compared with the associated Doppler spectral width and peak Doppler frequency. Results show that Doppler spectral width, decorrelation time, and the standard deviation of Doppler spectra are correlated. In addition, the Doppler frequency shift induced by the motion of the water surface is analyzed by peak Doppler frequency and Doppler centroid, which show dependence on radar geometry and wind direction.

Description: Hyperspectral images in remote sensing systems with rich spatial and spectral information provide an opportunity for researchers to discover the world. Anomaly detection is one of the most interesting topics over the last two decades in hyperspectral imagery (HSI). In this letter, we propose a modified collaborative-representation-based with outlier removal anomaly detector (CRBORAD) for anomaly detection. We use both spectral and spatial information for detecting anomalies since that is more precise than using only spectral information. The proposed detector can adaptively estimate the background by its adjacent pixels within a sliding dual-window. We remove outlier pixels that are significantly different from majority of pixels, before estimating background pixels. It can lead us to precise detection of anomalies in subsequent stages. By subtracting the predicted background from the original HSI, the residual image is resulted and anomalies can be determined, finally. Kernel extension of the proposed approach is also presented. CRBORAD results on San Diego airport and the Rochester Institute of Technology data are illustrated using intuitive images, receiver operating characteristic curves, and area under curve values. The results are compared with four popular and previous methods and prove the superiority of the proposed CRBORAD method.

Description: Recently, collaborative representation has received much attention in the hyperspectral image (HSI) classification due to its simplicity and effectiveness. However, the existing collaborative representation-based HSI classification methods ignore the correlation among different classes. To overcome this problem, we propose a discriminative kernel collaborative representation and Tikhonov regularization method (DKCRT) for HSI classification, which can make the kernel collaborative representation of different classes to be more discriminative. Specifically, the kernel trick is adopted to map the original HSI into a high space to improve the class separability. Besides, distance-weighted kernel Tikhonov regularization is adopted to enforce these training samples to have large representation coefficients, which are similar to the test sample in the high-dimensional feature space. Moreover, we add a discriminative regularization term to further enhance the separability of different classes, which can take the correlation among different classes into consideration. Furthermore, to take the spatial information of HSI into consideration, we extend the DKCRT to a joint version named JDKCRT. Experiments on real HSIs demonstrate the efficiency of the proposed DKCRT and JDKCRT.

Description: Retrieved soil moisture estimates from the National Aeronautics and Space Administration (NASA) Soil Moisture Active Passive (SMAP) radiometer are assimilated into the Noah land surface model (LSM) within the NASA Land Information System. The experimental testbed is based on a real-time LSM system produced by the NASA Short-Term Prediction Research and Transition Center. A nonlocalized cumulative distribution function-matching bias correction (BC) is applied to the SMAP retrievals, with separate correction curves calculated based on soil texture categories. We show that the assimilation of SMAP soil moisture retrievals with nonlocalized BC can mitigate two types of artifacts due to spatially varying errors in the forcing data from: 1) bad point (rain gauge) data and 2) strong gradients along the eastern U.S.–Canada border, resulting from blending different observing systems.

Description: The hardware instability of a ground penetrating radar (GPR) system has a severe impact on the quantitative analysis of GPR data, which is aimed for material characterization and subsurface monitoring. In this letter, an instability index is proposed to quantify the stability performance of a GPR system and the influences of the GPR system type, warm-up time, environmental noise, and the antenna vibration on it are evaluated through a series of laboratory experiments on a sandbox model. It is found that the GPR signal recorded by a stepped-frequency GPR system based on a vector network analyzer is much more stable than that by a commercial impulse GPR system at a cost of more sweep time. A warm-up time of several minutes is enough for an impulse GPR system. Environmental noise has a negligible influence on the stability performance of a GPR system. Mechanical vibrations of GPR antennas have a severe impact on the stability performance of the GPR system, and the instability index and timing jitter can be increased by more than one order of magnitude in a vibrating condition over those in a static condition. The instability index of the direct signal has a negligible difference with that of the reflection signal from a metal plate; thus, a simple measurement of direct signal on the ground surface is suggested for the evaluation of the instability of a GPR system in field in the future.

Description: This letter proposes a new approach for knowledge-aided estimation of structured clutter covariance matrices (CCMs) in airborne radar systems with limited training data. First, we model the CCM in space–time adaptive processing (STAP) as a sum of low-rank Kronecker products. We then apply a permutation operation to convert the Kronecker factors into linear structures and propose a novel CCM estimation method under the maximum-likelihood framework. Employing a proximal gradient algorithm, the proposed method simultaneously exploits the knowledge about the clutter and the Kronecker structure of the CCM. We finally evaluate the performance of the proposed method using real data from airborne STAP.

Description: Water-body segmentation is an important issue in remote sensing and image interpretation. Classic methods for counteracting this problem usually include the construction of index features by combining different spectra, however, these methods are essentially rule-based and fail to take advantage of context information. Additionally, as the quality of image resolution improves, these methods are proved to be inadequate. With the rise of convolutional neural networks (CNN), the level of research about segmentation has taken a huge leap, but the field is still facing an increasing demand for data and the problem of blurring boundaries. In this letter, a new segmentation network called restricted receptive field deconvolution network (RRF DeconvNet) is proposed, with which to extract water bodies from high-resolution remote sensing images. Compared with natural images, remote sensing images have a weaker pixel neighborhood relativity; in consideration of this challenge, an RRF DeconvNet compresses the redundant layers in the original DeconvNet and no longer relies on a pretrained model. In addition, to tackle the blurring boundaries that occur in CNN, a new loss function called edges weighting loss is proposed to train segmentation networks, which has been shown to significantly sharpen the segmentation boundaries in results. Experiments, based on Google Earth images for water-body segmentation, are presented in this letter to prove our method.

Description: Infrared (IR) small target detection with high detection rate, low false alarm rate, and high detection speed has a significant value, but it is usually very difficult since the small targets are usually very dim and may be easily drowned in different types of interferences. Current algorithms cannot effectively enhance real targets and suppress all the types of interferences simultaneously. In this letter, a multiscale detection algorithm utilizing the relative local contrast measure (RLCM) is proposed. It has a simple structure: first, the multiscale RLCM is calculated for each pixel of the raw IR image to enhance real targets and suppress all the types of interferences simultaneously; then, an adaptive threshold is applied to extract real targets. Experimental results show that the proposed algorithm can deal with different sizes of small targets under complex backgrounds and has a better effectiveness and robustness against existing algorithms. Besides, the proposed algorithm has the potential of parallel processing, which is very useful for improving the detection speed.

Description: Advertisements.

Description: Provides instructions and guidelines to prospective authors who wish to submit manuscripts.

Description: Today, very dense synthetic aperture radar (SAR) time series are available through the framework of the European Copernicus Programme. These time series require innovative processing and preprocessing approaches including novel speckle suppression algorithms. Here we propose an image transform for hypertemporal SAR image time stacks. This proposed image transform relies on the temporal patterns only, and therefore fully preserves the spatial resolution. Specifically, we explore the potential of empirical mode decomposition (EMD), a data-driven approach to decompose the temporal signal into components of different frequencies. Based on the assumption that the high-frequency components are corresponding to speckle, these effects can be isolated and removed. We assessed the speckle filtering performance of the transform using hypertemporal Sentinel-1 data acquired over central Germany comprising 53 scenes. We investigated speckle suppression, ratio images, and edge preservation. For the latter, a novel approach was developed. Our findings suggest that EMD features speckle suppression capabilities similar to that of the Quegan filter while preserving the original image resolution.

Description: Cross-scene hyperspectral image (HSI) classification has recently become increasingly popular due to its crucial use in various applications. It poses great challenges to existing domain adaptation methods because of the data set shift, that is, two scenes exhibit huge distribution discrepancy. To tackle this problem, we propose a new domain adaptation method called hyperspectral feature adaptation and augmentation (HFAA) for cross-scene HSI classification. The proposed HFAA method learns a common subspace by introducing two different projection matrices to extract the transferable knowledge from the source domain to the target domain. To further enhance the common subspace representation, we propose to augment it by the feature selection strategy. HFAA can make full use of the original features from both source and target domains, and increase the similarity of the samples with the same label from the two domains. Our proposed HFAA method achieves compact but discriminative feature representations, which make it well suited for data sets with a large number of classes and huge interclass ambiguity. Experimental results on the Earth Observing 1 hyperspectral data set show that HFAA can produce state-of-the-art performance and surpass previous methods.

Description: In this letter, the recently developed extreme gradient boosting (Xgboost) classifier is implemented in a very high resolution (VHR) object-based urban land use–land cover application. In detail, we investigated the sensitivity of Xgboost to various sample sizes, as well as to feature selection (FS) by applying a standard technique, correlation-based FS. We compared Xgboost with benchmark classifiers such as random forest (RF) and support vector machines (SVMs). The methods are applied to VHR imagery of two sub-Saharan cities of Dakar and Ouagadougou and the village of Vaihingen, Germany. The results demonstrate that Xgboost parameterized with a Bayesian procedure, systematically outperformed RF and SVM, mainly in larger sample sizes.

Description: Polarimetric synthetic aperture radar (PolSAR) image classification is an important application. Advanced deep learning techniques represented by deep convolutional neural network (CNN) have been utilized to enhance the classification performance. One current challenge is how to adapt deep CNN classifier for PolSAR classification with limited training samples, while keeping good generalization performance. This letter attempts to contribute to this problem. The core idea is to incorporate expert knowledge of target scattering mechanism interpretation and polarimetric feature mining to assist deep CNN classifier training and improve the final classification performance. A polarimetric-feature-driven deep CNN classification scheme is established. Both classical roll-invariant polarimetric features and hidden polarimetric features in the rotation domain are used to drive the proposed deep CNN model. Comparison studies validate the efficiency and superiority of the proposal. For the benchmark AIRSAR data, the proposed method achieves the state-of-the-art classification accuracy. Meanwhile, the convergence speed from the proposed polarimetric-feature-driven CNN approach is about 2.3 times faster than the normal CNN method. For multitemporal UAVSAR data sets, the proposed scheme achieves comparably high classification accuracy as the normal CNN method for train-used temporal data, while for train-not-used data it obtains an average of 4.86% higher overall accuracy than the normal CNN method. Furthermore, the proposed strategy can also produce very promising classification accuracy even with very limited training samples.

Description: Hyperspectral remote sensing image (HSI) clustering can be defined as the process of segmenting pixels into different sets that satisfy the requirement that the differences between sets are much greater than the differences within sets. According to the fast density peak-based clustering algorithm, we propose an unsupervised HSI clustering method based on the density of pixels in the spectral space and the distance between pixels. For the metric of the density, we present an adaptive-bandwidth probability density function using pixel numbers as the input and the calculated pixel local density as the output, which determines the bandwidth on the basis of the Gaussian assumption. For the metric of the distance, in order to obtain a pixel-level spectral distance, we calculate the Euclidean distance between pixel vectors from the multiple bands. In the proposed approach: 1) use the least-squares method for the curve fitting of the two results; 2) eliminate outliers based on the Pauta criterion; 3) adopt regression calculation; and 4) obtain the cluster centers according to the classification criteria of the local density and the distance between pixel vectors. The other noncluster center points are clustered based on their similarities with the cluster centers by iteration. Finally, we compare the results with those of other unsupervised clustering methods and the reference data sets.

Description: Atmospheric compensation (AC) is a fundamental and critical step for quantitative exploitation of hyperspectral data. It is the means by which the reflectance of an object/material is estimated from the measured at-sensor radiance. Such reflectance is the inherent signature that is used to identify various materials in a monitored scene. AC is quite complex and is hampered by the large amount of uncontrollable variables that play a role: just think about the spatial variability of some atmospheric constituents such as water vapor and aerosols, or to the rapidly spatially varying effects of the radiation coming from adjacent areas. Though, in principle, some atmospheric parameters and radiometric quantities such as solar irradiance and sky irradiance can be measured during the flight, in practice such measures are rarely available in an operational framework or are taken at a single point of the surface ignoring their spatial variation. Thus, a prompt quantitative exploitation of hyperspectral data for operational purposes, such as material identification and object detection, requires unsupervised and accurate AC procedures that can learn from the image itself the parameters of the inversion model and follow their variability within the scene. In this framework, we present a new unsupervised methodology for AC of airborne hyperspectral images in the visible and near-infrared spectral range. The proposed methodology relies on a radiative transfer model accounting for the adjacency effect and allows the estimation of relevant atmospheric parameters. Specifically, it embeds two new algorithms for the estimation of: 1) aerosol and atmospheric visibility and 2) the water vapor content of the atmosphere accounting for the spatial variability of such a parameter. The two algorithms significantly differ from those adopted by existing state-of-the-art approaches or in commercial packages such as fast line-of-sight atmospheric analysis of spectral hypercubes and airborne atm- spheric and topographic correction algorithm. In this paper, we present the detailed description of the new AC methodology, and we analyze the results provided by the algorithm over real data.

Description: Sparsity-regularized synthetic aperture radar (SAR) imaging framework has shown its remarkable performance to generate a feature-enhanced high-resolution image, in which a sparsity-inducing regularizer is involved by exploiting the sparsity priors of some visual features in the underlying image. However, since the simple prior of low-level features is insufficient to describe different semantic contents in the image, this type of regularizer will be incapable of distinguishing between the target of interest and unconcerned background clutters. As a consequence, the features belonging to the target and clutters are simultaneously affected in the generated image without concerning their underlying semantic labels. To address this problem, we propose a novel semantic information guided generative framework for target-oriented SAR image formation, which aims at enhancing the interested target scatters while suppressing the background clutters. First, we develop a new semantics-specific regularizer for image formation by exploiting the statistical properties of different semantic categories in a target scene SAR image. In order to infer the semantic label for each pixel in an unsupervised way, we moreover induce a novel high-level prior-driven regularizer and some semantic causal rules from the prior knowledge. Finally, our regularized framework for image formation is further derived as a simple iteratively reweighted $ell _{1}$ minimization problem that can be conveniently solved by many off-the-shelf solvers. Experimental results demonstrate the effectiveness and superiority of our framework for SAR image formation in terms of target enhancement and clutters suppression, compared with the state of the arts. Additionally, the proposed framework opens a new direction of devoting some machine learning strategies to image formation, which can benefit the subsequent decision-making tasks.

Description: Small-scale glaciological processes can drive large-scale ice sheet behavior but remain underreported due to a paucity of surface elevation measurements in remote polar regions. Satellite images provide a relatively long record of spatially dense surface observations, which allow us to investigate changes in ice sheet topography on the spatial scales of 1–10 km. Inferring surface topography from satellite images is an established technique, but in previous efforts, strict requirements for illumination conditions and image quality have led to a great quantity of discarded data. Relaxing quality requirements and fitting linear trends to the time series of image-derived surface topography allow inclusion of more total signal and enable blending data from multiple platforms. As a proof of concept, we combine 121 MODerate-resolution Imaging Spectroradiometer images to develop a 250-m resolution map of surface elevation change at Totten Glacier, Antarctica achieving a 1- $sigma $ uncertainty of 0.22 $textrm {m},textrm {a}^{-1}$ for a 15-year period. Our method of repeat photoclinometry agrees with repeat laser altimetry while revealing clear patterns of surface elevation change associated with ice advection, channelized ice shelf basal melt, subglacial lake activity, and possible grounding line migration.

Description: Amplitude-versus-offset (AVO) inversion always plays an important role in reservoir fluid identification, which allows the estimation of various rock and fluid properties from prestack seismic data. In this paper, we propose a new method for discrimination of hydrocarbon accumulation that combines frequency-dependent AVO inversion scheme and variational mode decomposition (VMD). VMD is a recently developed algorithm for adaptive signal decomposition that is able to nonrecursively decompose a multicomponent signal into a number of quasi-orthogonal intrinsic mode functions and avoid mode mixing effectively. VMD is superior to other state-of-the-art approaches in obtaining high-resolution and high-fidelity local time–frequency depiction performance. Two synthetic signals are employed to illustrate that VMD achieves higher temporal and frequency resolution than the conventional continuous wavelet transform (CWT) decomposition. Other synthetic examples, elastic and dispersive, are utilized to demonstrate that the proposed method is more reliable for the detection of hydrocarbon saturation and a comparison is made with the CWT-based inverted results. Application on field data has further shown that the proposed approach has the potential in identifying the reservoir related to hydrocarbon.

Description: The low-level temperature inversions have significant impacts on Arctic climate change feedbacks. The Atmospheric Infrared Sounder (AIRS) can extract the inversions over both land and ocean, and it is, however, sensitive to the presence of clouds. In this paper, we evaluate the effect of cloud fraction (CF) on AIRS inversions over both land and ocean. First, the AIRS inversions under clear-sky conditions are compared with the results from the microwave-based Constellation Observing System for Meteorology, Ionosphere, and Climate (COSMIC) observations in 2007–2013. Results show that despite the COSMIC and AIRS inversions appearing to be deeper and stronger in winter than in autumn, spring, and summer, the former is generally shallower and stronger than the latter in all seasons over both land and ocean. Time-series analysis of the mean monthly inversions from COSMIC and AIRS observations in 2007–2013 under both clear-sky and cloudy conditions further indicates that their differences are systematic and can be effectively mitigated after calibration under all sky conditions. Taking the calibrated COSMIC inversions as references, the AIRS inversion depths can be estimated with a root mean square (rms) of less than about 86 and 135 m, and the AIRS inversion strength can be obtained with an rms of better than about 1.7 °C and 1.3 °C under cloudy conditions over land and ocean, respectively. Moreover, while the AIRS inversion depths are insensitive to CF variations over both land and ocean, the inversion strengths are more sensitive to the CF variations over land than ocean.

Description: Hyperspectral image classification has become a research focus in recent literature. However, well-designed features are still open issues that impact on the performance of classifiers. In this paper, a novel supervised deep feature extraction method based on siamese convolutional neural network (S-CNN) is proposed to improve the performance of hyperspectral image classification. First, a CNN with five layers is designed to directly extract deep features from hyperspectral cube, where the CNN can be intended as a nonlinear transformation function. Then, the siamese network composed by two CNNs is trained to learn features that show a low intraclass and high interclass variability. The important characteristic of the presented approach is that the S-CNN is supervised with a margin ranking loss function, which can extract more discriminative features for classification tasks. To demonstrate the effectiveness of the proposed feature extraction method, the features extracted from three widely used hyperspectral data sets are fed into a linear support vector machine (SVM) classifier. The experimental results demonstrate that the proposed feature extraction method in conjunction with a linear SVM classifier can obtain better classification performance than that of the conventional methods.

Description: The fraction of absorbed photosynthetically active radiation (FAPAR) is a critical input in many climate and ecological models. The accuracy of satellite FAPAR products directly influences estimates of ecosystem productivity and carbon stocks. The targeted accuracy of FAPAR products is 10% or 0.05 for many applications. However, most current FAPAR products do not meet such requirements, and further improvements are still needed. In this paper, a data fusion scheme based on the multiple resolution tree (MRT) approach is developed to integrate multiple satellite FAPAR estimates at site and regional scales. MRT was chosen because of the superior computational efficiency compared with other fusion methods. The fusion scheme removed the bias in FAPAR estimates and resulted in a 15% increase in the $R^{2}$ and 3% reduction in the root-mean-square error compared with the average of individual FAPAR estimates. The regional-scale fusion filled in the missing values, and provided spatially consistent FAPAR distributions at different resolutions. Overall, MRT can be used to efficiently and accurately generate spatially and temporally continuous FAPAR data across both site and regional scales.

Description: Synthetic aperture radar (SAR) image classification is a fundamental process for SAR image understanding and interpretation. With the advancement of imaging techniques, it permits to produce higher resolution SAR data and extend data amount. Therefore, intelligent algorithms for high-resolution SAR image classification are demanded. Inspired by deep learning technology, an end-to-end classification model from the original SAR image to final classification map is developed to automatically extract features and conduct classification, which is named deep recurrent encoding neural networks (DRENNs). In our proposed framework, a spatial feature learning network based on long–short-term memory (LSTM) is developed to extract contextual dependencies of SAR images, where 2-D image patches are transformed into 1-D sequences and imported into LSTM to learn the latent spatial correlations. After LSTM, nonnegative and Fisher constrained autoencoders (NFCAEs) are proposed to improve the discrimination of features and conduct final classification, where nonnegative constraint and Fisher constraint are developed in each autoencoder to restrict the training of the network. The whole DRENN not only combines the spatial feature learning power of LSTM but also utilizes the discriminative representation ability of our NFCAE to improve the classification performance. The experimental results tested on three SAR images demonstrate that the proposed DRENN is able to learn effective feature representations from SAR images and produce competitive classification accuracies to other related approaches.

Description: We present a new synthetic aperture radar (SAR) raw signal simulator, which is able to simultaneously generate raw signals of different polarimetric channels of a polarimetric SAR system in such a way that a correct covariance matrix is obtained for the final images. Extended natural scenes, dominated by surface scattering, are considered. A fast Fourier-domain approach is used for the generation of raw signals. Presentation of theory is supplemented by meaningful experimental results, including a comparison of simulations with real polarimetric scattering data.

Description: Sparse representation has been widely used in the field of remote sensing image super-resolution (SR) to restore a high-quality image from a low-resolution (LR) image, e.g., from the blurred and downsampled version of an LR image’s high-resolution (HR) counterpart. It is well known that each image patch can be represented by a linear combination of the atoms of an overcomplete dictionary, and we can obtain an expression of sparse coefficients by $l_{1}$ norm regularization. Owing to the lack of an inner relationship between image patches and an image’s global information, the traditional methods of jointly training two overcomplete dictionaries cannot obtain good SR results. Therefore, we propose an effective approach for remote sensing image SR based on sparse representation. More specifically, a novel global joint dictionary model (GJDM) is used to explore the prior knowledge of images, including local and global characteristics. First, we train two dictionaries for detail image patches and HR patches. Second, in order to enhance the inner relationship between image patches, we introduce a global self-compatibility model for global regularization. Finally, the sparse representation and the local and nonlocal constraints are integrated to improve the performance of the model, and the fast adaptive shrinkage-thresholding algorithm is employed to solve the convex optimization problem in the GJDM. Compared with other methods, the results of the proposed method show good SR performance in preserving details and texture information and significant improvement in a peak signal-to-noise ratio.

Description: This paper presents the estimation of the ground elevation and canopy top elevation from the data collected by an airborne frequency-modulated continuous waveform profiling radar, Tomoradar. The estimated ground and canopy top elevations are critical for the derivation of reference information for the satellite-borne microwave radar data and the modeling of interaction between microwave radar signal and foliage. The methods of estimating the ground elevation and canopy top elevation from profiling radar are introduced, and the accuracy was evaluated via digital terrain model and Velodyne VLP-16 LiDAR integrated with the Tomoradar. To our knowledge, the ranging radar and the LiDAR data were simultaneously collected for the first time. The evaluation proved that the root-mean-square error (RMSE) of ground level estimation of the developed profiling radar can reach up to 0.33 m. When comparing the estimated canopy top peak elevation between the profile radar data and the LiDAR data, it was found that the side lobes of Tomoradar antenna system may produce undesired canopy backscatters when the size of the canopy gap is comparable to the footprint size of the main lobe, resulting in a higher canopy top elevation measurement from Tomoradar than that from LiDAR. The RMSE of the estimated canopy top peak elevation between two data sets was 0.32 m in the best case and 0.852 m on average. Moreover, the RMSE of point-to-point comparing the entire canopy tops elevation estimated from the data of two active remote sensing systems is 0.799 m after excluding the outliers.

Description: For airborne repeat-pass synthetic aperture radar interferometry (InSAR), precise trajectory information is needed to compensate for deviations of the platform movement from a linear track. Using the trajectory information, motion compensation (MoCo) can be implemented within SAR data focusing. Due to the inaccuracy of current navigation systems, residual motion errors (RMEs) exist between the real and measured trajectory, causing phase undulations in the final interferograms. Up to now, MoCo and RME estimation have usually been combined in airborne InSAR to estimate ground deformation. Conventional MoCo methods generally involve azimuthal and range resampling and phase correction. Then frequency-domain focusing techniques can be used to generate the SAR images. After focusing SAR images with MoCo, both multisquint and autofocus approaches can be used to estimate RME. In addition to the MoCo-based frequency-domain focusing, the time-domain backprojection (BP) technique can also focus the SAR data obtained from highly nonlinear platform trajectories. In this paper, we present, for the first time, the combination of BP and multisquint techniques for RME estimation. A detailed derivation of the implementation of the multisquint approach using the BP-focusing images is presented. Repeat-pass data from the SlimSAR system over Slumgullion landslide are used to demonstrate the feasibility of RME estimation for both stationary and nonstationary scenes. We conclude that the proposed method can effectively remove the RME.

Description: Transform sensing is proposed to sense the space through certain beam patterns designed for a transformation basis. The received signals will be the result of transformation instead of the original raw data. In doing so, the process of generating an arbitrary beam pattern on the phased array and multiple-input multiple-output is described. We further propose a transform-sensing mechanism using wavelets, and compare it with traditional sensing methods. Simulations and experiments demonstrate how to generate the transmission patterns and sense through the transform-sensing mechanism. The results show that transform sensing obtains high-resolution data on the target area, while spending less time on nontarget ones that need low resolutions. In this way, the new sensing mechanism generates a multiresolution result, which balances resolution and sensing efficiency.

Description: Focusing on open forests and woodlands within the Injune Landscape Collaborative Project research area in central southeast Queensland, Australia, and using dual-pol (HH and HV) ALOS PALSAR repeat-pass InSAR data (temporal baseline of 92 days), this paper explores the detection of forest disturbance from the spaceborne repeat-pass InSAR correlation magnitude by developing a simple and efficient forest disturbance detection approach. In particular, a generic physical InSAR scattering model is derived by accounting for the forest disturbance information as well as the normal temporal decorrelation effects that are later compensated for using the modified Random Volume over Ground model. Based on the generic model, a quantitative indicator of forest disturbance is retrieved, namely, disturbance index that varies from 0 (no disturbance) to 1 (complete deforestation). This index is compared with that identified using a time series of Landsat sensor data over a selective logging area and has a relative root mean square error of 13% at a spatial resolution of 0.8 ha. This paper highlights the use of the co-pol InSAR correlation magnitude for forest disturbance detection, which serves as a complimentary application to using the cross-pol counterpart for forest height inversion in a companion work. Given the global availability of this type of data (e.g., Japanese Aerospace Exploration Agency’s ALOS-1/2 and NASA-ISRO’s NISAR), the method is anticipated to contribute to the range of tools being developed for large-scale forest disturbance assessment and monitoring.

Description: Robust and effective detection of small target and false alarm (FA) suppression are the key techniques in infrared search and track systems. In this paper, the derivative entropy-based contrast measure (DECM) is proposed for small-target detection under various complex background clutters. Initially, different directional derivatives of an infrared image are calculated based on the facet model. Then, by analyzing the derivative properties of small target, the primitive entropy formula is improved by incorporating derivative information. With the improved entropy, the contrast measure is constructed to enhance small target and suppress background clutters in each derivative subband. Finally, the contrast measure maps derived from derivative subbands are fused together. The small target could be segmented easily from the fusion result. Experimental results demonstrate that DECM could effectively enhance dim small targets and suppress complex background clutters. Besides, DECM is also robust to infrared small-target images with noises of different levels. The detection results achieve higher detection ratio and lower FA compared with those of other methods under various infrared scenes.

Description: Advertisement, IEEE. IEEE Collabratec is a new, integrated online community where IEEE members, researchers, authors, and technology professionals with similar fields of interest can network and collaborate, as well as create and manage content. Featuring a suite of powerful online networking and collaboration tools, IEEE Collabratec allows you to connect according to geographic location, technical interests, or career pursuits. You can also create and share a professional identity that showcases key accomplishments and participate in groups focused around mutual interests, actively learning from and contributing to knowledgeable communities. All in one place! Learn about IEEE Collabratec at ieeecollabratec.org.

Description: Provides instructions and guidelines to prospective authors who wish to submit manuscripts.

Description: Presents the front cover for this issue of the publication.

Description: The IEEE Geocience and Remote Sensing Society (GRSS) Nominations Committee calls upon our membership to nominate members to serve on the GRSS Administrative Committee (AdCom). A nominating petition carrying a minimum of 2% of the names of eligible Society members (~70), excluding students, shall automatically place that nominee on the slate. Such nominations must be made by 30 April 2018. The Nominations Committee may choose to include a name on the slate regardless of the number of names generated by the nominating petition process. Prior to submission of a nomination petition, the petitioner shall have determined that the nominee named in the petition is willing to serve if elected; and evidence of such willingness to serve shall be submitted with the petition. Candidates must be current Members of the IEEE and the GRSS. Petition signatures can be submitted electronically through the Society website or by signing, scanning, and electronically mailing the pdf file of the paper petition. The name of each member signing the paper petition shall be clearly printed or typed. For identification purposes of signatures on paper petitions, membership numbers and addresses as listed in the official IEEE Membership records shall be included. Only signatures submitted electronically through the Society website or original signatures on paper petitions shall be accepted. Further information is provided here.

Description: Presents the table of contents for this issue of the publication.

Description: Presents a listing of the editorial board, board of governors, current staff, committee members, and/or society editors for this issue of the publication.

Description: Presents the President’s message for this issue of the publication.

Description: Prospective authors are requested to submit new, unpublished manuscripts for inclusion in the upcoming event described in this call for papers.

Description: The Lombardy Aerospace Industry Cluster [1] was founded in Italy in 2014 as the final step in constructing a representative institution for the regional system of aerospace industries. The process was initiated in 2009 with the first formal contacts between the local industry federation and the regional government of Lombardy, which led to the foundation of a provisional body (Distretto Aerospaziale Lombardo) in preparation for a more formalized industry cluster to come and in light of a mandate to represent the local aerospace industry politically.

Description: Hyperspectral imagery contains hundreds of contiguous bands with a wealth of spectral signatures, making it possible to distinguish materials through subtle spectral discrepancies. Because these spectral bands are highly correlated, dimensionality reduction, as the name suggests, seeks to reduce data dimensionality without losing desirable information. This article reviews discriminant analysisbased dimensionality-reduction approaches for hyperspectral imagery, including typical linear discriminant analysis (LDA), state-of-the-art sparse graph-based discriminant analysis (SGDA), and their extensions.

Description: Presents the introductory editorial for this issue of the publication.

Description: Describes the above-named upcoming conference event. May include topics to be covered or calls for papers.

Description: Wind profilers (WPs) are ground-based pulsed Doppler radars operating in the UHF/very high frequency (VHF) bands. They use backscatter from clear-air turbulence as a tracer of background wind in the troposphere and up to mesosphere. Wind profiling systems range from large research radar systems to smaller operational units. In recent times, multiparameter cost function (MPCF) has emerged as a computationally efficient Doppler profile estimation method. It has been claimed that the MPCF method can be easily migrated to any vertically sounding clear-air wind profiling system that works in Doppler beam swinging mode. In order to investigate this claim, MPCF was applied to the following wind profiling systems: the middle and upper atmosphere (MU) radar, located at Shigaraki, Japan, which is an active phased array VHF radar system with 475 transceivers, and Doppler sodar, located at Pune, India, which uses acoustic frequency to obtain echoes in the planetary boundary layer. Both these systems are complementary in the sense that they cover ranges from ground to about 80 km. During the experimentation, the MPCF algorithm did not need any change of parameters except the matching of the data reading formats. The results of MPCF on the MU radar were validated with Radiosonde data. These results indicate that the MPCF works seamlessly on all types of WP systems irrespective of the carrier, range, and radar type.

Description: An efficient method based on the evolutionary programming (EP) technique is proposed for inverse profiling of 2-D buried dielectric objects with elliptical cross sections. In particular, EP with Cauchy mutation operator (EP-CMO), as its first reported implementation to inverse problems, is utilized as a stochastic optimization tool for quantitatively reconstructing buried objects. Moreover, the method of moments technique in conjunction with conjugate gradient-fast Fourier transform method is used, as a fast and simple frequency domain forward solver, in each iteration of the proposed method. Numerical results for different case studies are presented and analyzed. To assess the proposed EP-CMO method, the results are also compared statistically with that of three other well-known optimization techniques, namely, EP with Gaussian mutation, particle swarm optimization, and genetic algorithms. The results reveal that EP-CMO is a significantly more robust and efficient optimization tool in reconstruction of this class of buried objects.

Description: Magnetic anomaly detection is an effective approach for detecting the visually obscured ferromagnetic target, and its performance is mainly limited by background geomagnetic noise. In contrast to the traditional detection methods that rely on several a priori assumptions regarding the target or the probability of magnetic noise consisting of external geomagnetic noise and intrinsic sensor noise, we present, in this letter, a new estimator of information entropy for differential signal acquired by a pair of magnetic sensors to detect any changes in the magnetic noise pattern. First, the magnetic noise probability density function (PDF) of differential signal is estimated by using the kernel smoothing method. Then, the minimum entropy detector based on the magnetic noise PDF of differential signal is used to detect the magnetic anomaly target. Finally, according to the probabilities of false alarm, the detection threshold can be obtained to be used for abnormal judgment. In order to verify the effectiveness of the proposed method, the experiment is conducted, and the results demonstrate that the proposed method has better detection performance than that of traditional methods.

Description: This letter presents a band selection method relying on saliency bands and scale selection (SBSS). The SBSS method is used to excavate the hidden information of hyperspectral images effectively, while its underlying assumptions are: 1) it is reasonable to combine spectral and spatial information to excavate the intrinsic property of a hyperspectral image; 2) there are some saliency bands that can represent a hyperspectral image without significant information loss in data exploitation; and 3) saliency, scale, and image description have an intrinsic connection. The computational complexity of the SBSS method is linear, and experimental results demonstrate that the proposed method obtains competitively good results compared with other state-of-the-art band selection techniques, in terms of classification accuracy.

Description: The lack of proper class discrimination among the hyperspectral (HS) data points poses a potential challenge in HS classification. To address this issue, this letter proposes an optimal geometry-aware transformation for enhancing the classification accuracy. The underlying idea of this method is to obtain a linear projection matrix by solving a nonlinear objective function based on the intrinsic geometrical structure of the data. The objective function is constructed to quantify the discrimination between the points from dissimilar classes on the projected data space. Then, the obtained projection matrix is used to linearly map the data to more discriminative space. The effectiveness of the proposed transformation is illustrated with three benchmark real-world HS data sets. The experiments reveal that the classification and dimensionality reduction methods on the projected discriminative space outperform their counterpart in the original space.

Description: Low-altitude aerial photography using small unmanned aerial vehicles (SUAVs) with large viewpoint changes causes nonrigid distortions and low overlap ratios. We present a nonrigid feature-based low-altitude SUAV image-registration method. The key idea of our method is to maintain a high matching ratio on inliers while taking advantage of outliers for varying the warping grids. Thus, accurate image transformation over the overlapping areas as well as a good approximation of the real transformation over the nonoverlapping areas can be obtained. Experiments on feature matching and image registration are performed using 42 pairs of SUAV images. Our method exhibited a favorable performance as compared with four state-of-the-art methods, even with up to 80% outliers.

Description: Wind profilers (WPs) are coherent pulsed Doppler radars operating in the ultrahigh frequency and very high frequency bands. These atmospheric radars receive backscattered signals from the atmospheric and nonatmospheric targets. Atmospheric targets include clear air turbulence, precipitation, mesospheric turbulence, ionospheric D, E, F regions, and meteors, whereas nonatmospheric targets are objects like airplanes, birds, insects, hills, and so on. Modern WPs operate for long hours and often change radar operating parameters. Echoes from one beam direction are observed for 8 to 32 s and a set of Doppler power spectra is generated. WPs generate more than a hundred sets of Doppler power spectra per hour. These Doppler spectra often contain echoes from more than one target. The studies for atmospheric modeling and prediction require analysis of the backscattered signal. In order to facilitate systematic study, the data need to be classified and archived according to the atmospheric target type. Considering the large data volume consisting of echoes from different atmospheric targets, there is a need of an automated classification technique that segregates the Doppler power spectral data according to the types of atmospheric targets. The technique is expected to operate in real time with the data generation. This paper presents a spectral feature-based classification (SFBC) method for the classification of Doppler power spectra. This method associates three to four descriptor spectral features with each type of atmospheric target. The SFBC method classifies the data into a particular type of the atmospheric target if concurrent occurrence of descriptor features corresponding to that target type is observed. This paper presents the results indicating that the SFBC classifies the data from different WP radars with good accuracy. It also shows that the SFBC method is computationally simpler compared with other established classification methods.

Description: For the first time, the performance of the range imaging technique of the very high frequency band middle and upper atmosphere (MU) radar (Shigaraki observatory, Japan), when using frequency diversity, is assessed. This is done by the detection of unmanned aerial vehicles (UAVs) operated near the radar during the Shigaraki UAV radar experiment campaign, carried out from June 1, 2015 to June 14, 2015. During this campaign devoted to the measurements of fine-scale turbulence and stability in the lower troposphere using the turbulence sensors on the DataHawk UAV and the MU radar, the detection of the small UAV by the MU radar provided an excellent opportunity for taking stock of the range imaging technique in the presence of a single hard target. It was found that range imaging reproduces a faithful image of the aircraft position and its displacements with an excellent accuracy (of the order of ~10 m), giving extra credence to the thin echo layers and their vertical displacements generally observed from the range imaging technique in stably stratified conditions.

Description: It is a feasible and promising way to utilize deep neural networks to learn and extract valuable features from synthetic aperture radar (SAR) images for SAR automatic target recognition (ATR). However, it is too difficult to effectively train the deep neural networks with limited raw SAR images. In this paper, we propose a new approach to do SAR ATR, in which a multiview deep learning framework was employed. Based on the multiview SAR ATR pattern, we first present a flexible mean to generate adequate multiview SAR data, which can guarantee a large amount of inputs for network training without needing many raw SAR images. Then, a unique deep convolutional neural network containing a parallel network topology with multiple inputs is adopted. The features of input SAR images from different views will be learned by the proposed network layer by layer; meanwhile, the learned features from the distinct views are fused in different layers progressively. Therefore, the proposed framework is able to achieve a superior recognition performance, and requires only a small number of raw SAR images for network training samples generation. Experimental results have shown the superiority of the proposed framework based on the Moving and Stationary Target Acquisition and Recognition data set.

Description: Inspired by recent development of artificial satellite, remote sensing images have attracted extensive attention. Recently, notable progress has been made in scene classification and target detection. However, it is still not clear how to describe the remote sensing image content with accurate and concise sentences. In this paper, we investigate to describe the remote sensing images with accurate and flexible sentences. First, some annotated instructions are presented to better describe the remote sensing images considering the special characteristics of remote sensing images. Second, in order to exhaustively exploit the contents of remote sensing images, a large-scale aerial image data set is constructed for remote sensing image caption. Finally, a comprehensive review is presented on the proposed data set to fully advance the task of remote sensing caption. Extensive experiments on the proposed data set demonstrate that the content of the remote sensing image can be completely described by generating language descriptions. The data set is available at https://github.com/201528014227051/RSICD_optimal .

Description: A two-stage sparse structure representation algorithm which can preserve the manifold structure of the data is proposed for synthetic aperture radar target configuration recognition in this paper. Manifold structure of the data is preserved by two stages. In the training stage, taking advantage of both the sparse representation (SR) and manifold learning, local structure of the data is preserved in the reconstruction space, where SR-based recognition is realized. In the testing stage, two structure preserving factors based on the testing samples are embedded into the SR model to enhance structure preserving performance. The first one is constructed to preserve the local structure of the testing samples, which can guarantee the samples that are close to each other in the original space will also be close to each other in the sparse space. And the second one is established to preserve the distant structure of the testing samples, which can ensure the samples that are far from each other in the original space will also be far from each other in the sparse space. Manifold structure of the data is well captured and preserved by two stages. Experimental results on the moving and stationary target acquisition and recognition database demonstrate the effectiveness of the proposed algorithm.

Description: Due to the very high orbit, the apparent features of geosynchronous synthetic aperture radar (GEOSAR) are the curved trajectory and long integration time, which can lead to severe coupling between the azimuth and the range directions and, therefore, complicates the resolution evaluation. The traditional analytical approach based on the 2-D division may produce large resolution error, and the numerical approach may suffer from huge computation burden. Therefore, an analytical resolution evaluation approach for GEOSAR based on the local feature of the ambiguity function is studied in this paper. The proposed approach is validated with simulation data to be of high efficiency and accuracy. In addition, the proposed approach is also demonstrated to be capable of evaluating the resolution for other complex platforms, and of evaluating the 3-D resolution of a SAR system.

Description: The accurate reconstruction of areas obscured by clouds is among the most challenging topics for the remote sensing community since a significant percentage of images archived throughout the world are affected by cloud covers which make them not fully exploitable. The purpose of this paper is to propose new methods to recover missing data in multispectral images due to the presence of clouds by relying on a formulation based on an autoencoder (AE) neural network. We suppose that clouds are opaque and their detection is performed by dedicated algorithms. The AE in our methods aims at modeling the relationship between a given cloud-free image (source image) and a cloud-contaminated image (target image). In particular, two strategies are developed: the first one performs the mapping at a pixel level while the second one at a patch level to take profit from spatial contextual information. Moreover, in order to fix the problem of the hidden layer size, a new solution combining the minimum descriptive length criterion and a Pareto-like selection procedure is introduced. The results of experiments conducted on three different data sets are reported and discussed together with a comparison with reference techniques.

Description: Differential interferometry using ground-based radar systems permits to monitor displacements in natural terrain with high flexibility in location, time of acquisition, and revisit time. In combination with polarimetric imaging, discrimination of different scattering mechanisms present in a resolution cell can be obtained simultaneously with the estimation of surface displacement. In this paper, we present the preprocessing steps and the calibration procedure required to produce high-quality calibrated polarimetric single-look complex imagery with KAPRI, a new portable Ku-band polarimetric radar interferometer. The processing of KAPRI data into single look complex images is addressed, including the correction of beam squint and of azimuthal phase variations. A polarimetric calibration model adapted to the acquisition mode is presented and used to produce calibrated polarimetric covariance matrix data. The methods are validated by means of a scene containing five trihedral corner reflectors. Data preprocessing is assessed by analyzing the oversampled response of a corner reflector, and the polarimetric calibration quality is verified by computing polarimetric signatures and residual calibration parameters.