ALBERT

Description: Because the principle of mirrored interferometric aperture synthesis (MIAS) is different from that of the conventional interferometric aperture synthesis, the antenna array for 1-D MIAS should be redesigned. And in order to get a precise estimation of the cosine visibilities, the maximum rank of the transfer matrix should be set as the key constraint condition of the array optimization model, but this has not been clearly presented before. In this letter, the maximum rank of the transfer matrix is discussed and proved by the mathematical induction method. When the position of each antenna in the array is an even multiple of 0.5, the value for the maximum rank is proven to be $M-2$ , and when the position of each antenna is an odd multiple of 0.5, the value for the maximum rank is proven to be $M-1$ , where M is the number of the spatial frequencies provided by the array. This conclusion is significant for the array design of 1-D MIAS.

Description: Characteristic analysis of sea clutter is important in utilizing radar observations and detecting sea-surface targets. Real data signals are analyzed to determine the multifractal characteristics of sea clutter signals. Sea clutter is a nonlinear, nonstationary radar echo signal. A novel method that detects targets in sea clutter is proposed by completely utilizing the strengths of empirical mode decomposition (EMD) and combining it with multifractal characteristics. The EMD method is applied to decompose sea clutter signals into several intrinsic mode functions (IMFs). Multifractal detrended fluctuation analysis is utilized to calculate the generalized Hurst exponent for the main functions of IMF after which real sea clutter data are used for training and testing. Results show that targets in sea clutter can be effectively observed and detected through the proposed method, the performance of which is better than that of the target detection method for the generalized Hurst exponent under typical time, fractional Fourier transform and wavelet transform domains.

Description: It is time consuming to annotate unlabeled remote sensing images. One strategy is taking the labeled remote sensing images from another domain as training samples, and the target remote sensing labels are predicted by supervised classification. However, this may lead to negative transfer due to the distribution difference between the two domains. To address this issue, we propose a novel domain adaptation method through transferring the parameters of extreme learning machine (ELM). The core of this method is learning a transformation to map the target ELM parameters to the source, making the classifier parameters of the target domain maximally aligned with the source. Our method has several advantages which was previously unavailable within a single method: multiclass adaptation through parameter transferring, learning the final classifier and transformation simultaneously, and avoiding negative transfer. We perform experiments on three data sets that indicate improved accuracy and computational advantages compared to baseline approaches.

Description: This letter develops a progressive band processing (PBP) of fast iterative pixel purity index (FIPPI) according to a band sequential acquisition format in such a way that FIPPI can be processed band by band, while band acquisition is ongoing. As a result, PBP-FIPPI can generate progressive profiles of interband changes among PPI counts which allow users to observe significant bands that capture PPI counts. The idea to implement PBP-FIPPI is to use an inner loop specified by skewers and an outer loop specified by bands to process FIPPI. Interestingly, these two loops can also be interchanged with an inner loop specified by bands and an outer loop iterated by growing skewers. The resulting FIPPI is called progressive skewer processing of FIPPI. It turns out that both versions provide different insights into the design of FIPPI.

Description: Multitemporal polarimetric synthetic aperture radar (SAR) data can be used to estimate the dominant scattering mechanism of targets in a stack of SAR data and to improve the performance of SAR interferometric methods for deformation studies. In this letter, we developed a polarimetric form of amplitude difference dispersion (ADD) criterion for time-series analysis of pixels in which interferometric noise shows negligible decorrelation in time and space in small baseline algorithm. The polarimetric form of ADD is then optimized in order to find the optimum scattering mechanism of the pixels, which in turn is used to produce new interferograms with better quality than single-pol SAR interferograms. The selected candidates are then combined with temporal coherency criterion for final phase stability analysis in full-resolution interferograms. Our experimental results derived from a data set of 17 dual polarizations X-band SAR images (HH/VV) acquired by TerraSAR-X shows that using optimum scattering mechanism in the small baseline method improves the number of pixel candidates for deformation analysis by about 2.5 times in comparison with the results obtained from single-channel SAR data. The number of final pixels increases by about 1.5 times in comparison with HH and VV in small baseline analysis. Comparison between persistent scatterer (PS) and small baseline methods shows that with regards to the number of pixels with optimum scattering mechanism, the small baseline algorithm detects 10% more pixels than PS in agricultural regions. In urban regions, however, the PS method identifies nearly 8% more coherent pixels than small baseline approach.

Description: Ground-based synthetic aperture radar (GBSAR) systems are popular remote sensing instruments for detecting ground changes of slopes, and small displacements of large structures as bridges, dams, and construction works. These radars are able to provide maps of displacement along range direction only. In this letter, we propose to use a transponder for operating a conventional linear GBSAR as a bistatic radar with the aim to acquire two different components of the displacement of the targets in the field of view.

Description: Data-driven classification algorithms have proved to do well for automatic target recognition (ATR) in synthetic aperture radar (SAR) data. Collecting data sets suitable for these algorithms is a challenge in itself as it is difficult and expensive. Due to the lack of labeled data sets with real SAR images of sufficient size, simulated data play a big role in SAR ATR development, but the transferability of knowledge learned on simulated data to real data remains to be studied further. In this letter, we show the first study of Transfer Learning between a simulated data set and a set of real SAR images. The simulated data set is obtained by adding a simulated object radar reflectivity to a terrain model of individual point scatters, prior to focusing. Our results show that a Convolutional Neural Network (Convnet) pretrained on simulated data has a great advantage over a Convnet trained only on real data, especially when real data are sparse. The advantages of pretraining the models on simulated data show both in terms of faster convergence during the training phase and on the end accuracy when benchmarked on the Moving and Stationary Target Acquisition and Recognition data set. These results encourage SAR ATR development to continue the improvement of simulated data sets of greater size and complex scenarios in order to build robust algorithms for real life SAR ATR applications.

Description: Synthetic aperture radar (SAR) altimeters reduce the along-track footprint size exploiting the coherence of the transmitted pulses and achieve at the same time a noise reduction. Consequently, a large effort has been aimed at the formulation of theoretical models that apply to SAR altimeters, in order to fully exploit the improvement in spatial resolution obtained from the along-track aperture synthesis. This letter presents a novel semianalytical waveform model for SAR interferometric altimeters that preserves high accuracy even in the presence of mispointing. Starting from the waveform model proposed by Wingham et al. that provides a unified formulation for pulse-limited and SAR interferometric altimeters which can only be computed numerically, here, we describe a semianalytical approximation for small variations of the mispointing angles around an arbitrary combination of pitch and roll angles $(\hat {\mu },\hat {\theta })$ . The proposed semianalytical waveform model allows to reduce the high dimensionality of the model proposed by Wingham et al. and it has been proven to be accurate for variations of mispointing angles up to 0.4 deg around the $(\hat {\mu },\hat {\theta })$ . The performance of the proposed formulation has been evaluated on simulated data from Sentinel-6 configuration and on real data from CryoSat-2 SARin acquisitions over ocean.

Description: The interferometric phases in highly sloped terrain have the characteristics of large fringe density, narrow width, low correlation, and under-sampling. The local fringe frequ- ency (LFF) is a criterion to evaluate the trend and magnitude of the local terrain gradient and can be employed to improve the quality of interferograms. The results of the traditional LFF estimation method can be affected by phase noise, and sometimes the phase unwrapping (PU) operation is also required for some local regions. When it comes to highly sloped terrain, the phenomenon of phase under-sampling may cause incorrectness in the absolute interferometric phase during the operation of PU and may then influence the accuracy of the whole estimation. In order to solve this problem, this letter proposes an extended maximum-likelihood method for LFF estimation based on the multifrequency interferometric synthetic aperture radar (InSAR) data. Through the differences in the LFF between the different frequency InSAR data, the estimation quality map is introduced to modify the large error in certain regions by local 2-D fitting and thus achieves a accurate estimation of LFF in highly sloped terrain. Finally, the estimated results of LFF are used to guide the process of phase filtering. Simulated data and real airborne dual-frequency InSAR data are both employed to validate this proposed method.

Description: High-resolution satellite imagery has been increasingly used on remote sensing classification problems. One of the main factors is the availability of this kind of data. Despite the high availability, very little effort has been placed on the zebra crossing classification problem. In this letter, crowdsourcing systems are exploited in order to enable the automatic acquisition and annotation of a large-scale satellite imagery database for crosswalks related tasks. Then, this data set is used to train deep-learning-based models in order to accurately classify satellite images that contain or not contain zebra crossings. A novel data set with more than 240000 images from 3 continents, 9 countries, and more than 20 cities was used in the experiments. The experimental results showed that freely available crowdsourcing data can be used to accurately (97.11%) train robust models to perform crosswalk classification on a global scale.

Description: The problem of synthetic aperture radar image recovery in the presence of wideband interference (WBI) is investigated. Delayed versions of a transmitted signal are utilized to construct a dictionary in which a signal of interest (SOI) has a sparse representation. In this letter, WBI is sparsely represented by the time-frequency domain. By utilizing the transform domains, a joint estimation approach is devised to simultaneously perform WBI suppression and SOI recovery within an optimization framework. Based on the separability property in the optimization, an alternating direction method of multipliers-based approach is developed to efficiently obtain a solution. Finally, simulation results are presented to demonstrate the superior performance of the joint estimation algorithm.

Description: Deep convolutional neural networks (DCNNs) have recently emerged as the highest performing approach for a number of image classification applications, including automated land cover classification of high-resolution remote-sensing imagery. In this letter, we investigate a variety of fusion techniques to blend multiple DCNN land cover classifiers into a single aggregate classifier. While feature-level fusion is widely used with deep neural networks, our approach instead focuses on fusion at the classification/information level. Herein, we train three different DCNNs: CaffeNet, GoogLeNet, and ResNet50. The effectiveness of various information fusion methods, including voting, weighted averages, and fuzzy integrals, is then evaluated. In particular, we used DCNN cross-validation results for the input densities of fuzzy integrals followed by evolutionary optimization. This novel approach produces the state-of-the-art classification results up to 99.3% for the UC Merced data set and the 99.2% for the RSD data set.

Description: The conflict between range and azimuth ambiguities is a challenging problem for spaceborne high-resolution and wide-swath (HRWS) synthetic aperture radar (SAR) imaging. In this letter, a novel ambiguity resolution approach based on frequency diverse array (FDA) is proposed to retrieve unambiguous signal from that with ambiguities in both range and azimuth domains. By exploiting the range-angle-dependent property of transmit steering vector in FDA and applying the second range dependence compensation approach, echoes from different ambiguous regions are separable in transmit spatial-Doppler frequency domains. Then a corresponding transmit beamformer is designed to extract spectrum component of each ambiguous region, which can be rearranged to comprise the desired unambiguous signal. Compatible with most existing azimuth ambiguity suppression algorithms which employ the receive degrees of freedom, the proposed approach can further enhance the capability of resolving ambiguity for HRWS SAR systems, and its effectiveness is verified by simulation experiments.

Description: Geographical information systems-based methods can be handled as powerful tools in assessing and quantifying impacts and, thus, supporting strategies for disaster risk reduction (DRR). This is particularly relevant on scenarios of global climate change and intensified increased human interventions on riverine systems. The Madeira River in Porto Velho city (Brazilian Amazon) is a good example of susceptible area to both of these factors. We take advantage of the 2014 flood, the largest recorded for this region, for combining remote sensing and geographic information system with socio, health, and infrastructure data to quantify spatially the flood impacts. Using high resolution airborne images, we applied a machine learning classification algorithm for detecting urban areas. Our results show that at the flood extent related to the highest river level at least 0.65 $\text {km}^{2}$ of urban area, 87 km of urban streets, four public schools, and two public health units were affected. More than 16 800 people suffered the impacts directly, and children represented 29.7% of them. Based on registered data, it was quantified that the city registered more than 20 cases of leptospirosis and the truck flow on the region decreased up to 92%. The spatially-explicit results of this letter are potential to guide strategies aiming to support decision-making for DRR.

Description: Spikelike noise is a common type of random noise existing in many geoscience and remote sensing data sets. The attenuation of spike-like noise has become extremely important recently, because it is the main bottleneck when processing the simultaneous source data that are generated from the modern seismic acquisition. In this letter, we propose a novel low-rank decomposition algorithm that is effective in rejecting the spike-like noise in the seismic data set. The specialty of the low-rank decomposition algorithm is that it is applied along the morphological direction of the seismic data sets with a prior knowledge of the morphology of the seismic data, which we call local slope. The seismic data are of much lower rank along the morphological direction than along the space direction. The morphology of the seismic data (local slope) is obtained via a robust plane-wave destruction method. We use two simulated field data examples to illustrate the algorithm workflow and its effective performance.

Description: Compact polarimetry (compact-pol), as an effective polarization system, can reduce the system complexity and data volume in comparison with quad polarimetry (quad-pol). Reconstruction of quad-pol data from compact-pol has been discussed mostly based on iterative algorithms which make use of the empirically parameterized model with the assumption of reflection symmetry of the scatterer. In this letter, a linear relationship between the compact-pol and quad-pol is first derived, and then the Wishart–Bayesian regularized inverse algorithm is developed to reconstruct pseudo quad-pol data from compact-pol. Such problem is solved using the efficient alternating direction method of multipliers to recover the pseudo quad-pol covariance matrix. The reconstruction performance is evaluated by coherence index, in comparison with existing methods.

Description: In this letter, we introduce a novel method for visual simultaneous localization and mapping (SLAM)—so-called Air-SSLAM—which exploits a stereo camera configuration. In contrast to monocular SLAM, scale definition and 3-D information are issues that can be more easily dealt with in stereo cameras. Air-SSLAM starts from computing keypoints and the correspondent descriptors over the pair of images, using good features-to-track and rotated-binary robust-independent elementary features, respectively. Then a map is created by matching each pair of right and left frames. The long-term map maintenance is continuously performed by analyzing the quality of each matching, as well as by inserting new keypoints into uncharted areas of the environment. Three main contributions can be highlighted in our method: 1) a novel method to match keypoints efficiently; 2) three quality indicators with the aim of speeding up the mapping process; and 3) map maintenance with uniform distribution performed by image zones. By using a drone equipped with a stereo camera, flying indoor, the translational average error with respect to a marked ground truth was computed, demonstrating promising results.

Description: In several applications, the automatic identification of regions of interest in remote sensing images is based on the assessment of the similarity of associated time series, i.e., two regions are considered as belonging to the same class if the patterns found in their spectral information observed over time are somewhat similar. In this letter, we investigate the use of a genetic programming (GP) framework to discover an effective combination of time series similarity functions to be used in remote sensing classification tasks. Performed experiments in a Forest–Savanna classification scenario demonstrated that the GP framework yields effective results when compared with the use of traditional widely used similarity functions in isolation.

Description: The recently proposed triplet Markov fields (TMF) model enhances the nonstationary image prior modeling ability by introducing an auxiliary field. Motivated by the TMF model, we propose a generalized TMF model based on ambiguity label information fusion (ALF-TMF) for synthetic aperture radar (SAR) image segmentation. The redefined auxiliary field in ALF-TMF indicates the dominant direction of local image contents and gives explicit nonstationary divisions of SAR images. To reduce the influence of unreliable observations caused by speckle noise, the original label field is adaptively generalized by introducing ambiguity class based on image observation and local nonstationary contextual information. Given the extended label field, prior and likelihood terms are constructed and merged to provide the posterior segmentation decision via the Bayesian fusion rule. Real SAR images are utilized in the experimental analysis, and the effectiveness of the proposed method is validated accordingly.

Description: The algorithm used in the retrieval of geophysical quantities from the Atmospheric Infrared Sounder (AIRS) instrument depends on two fundamental components. The first is a cost function that is the sum of squares of the differences between cloud-cleared radiances and their corresponding forward-model terms. The second is the minimization of this cost function. For the retrieval of carbon dioxide, the minimization is further improved using the method of vanishing partial derivatives (VPDs). In this letter, we show that this VPD component is identical to a coordinate descent method with Newton–Raphson updates, which allows it to be put in context with other optimization algorithms. We also show that the AIRS cost function is a limiting case of the cost function used in optimal estimation, which demonstrates how uncertainty quantification in the AIRS retrieval can be implemented.

Description: This letter addresses the problem of cross-range superresolution in Doppler beam sharpening (DBS). The coherence of echoes in the azimuth direction and the sparsity of the DBS image in the Doppler domain are fully exploited; thus, a superresolution DBS imaging framework using aperture-extrapolated sparse representation (SR) is proposed. In this framework, aperture extrapolation based on the autoregressive model is utilized to predict the forward and backward information in the azimuth direction, and SR is exploited to extract the Doppler spectrum information. In addition, the resolution ability with different coherent processing intervals is analyzed. The sharpening ratio in this proposed algorithm can be improved by a factor of two or four theoretically in comparison with the conventional DBS imaging method. Experimental results demonstrate that the proposed framework can lead to noticeable performance improvement.

Description: Spectral index time series can provide valuable phenological information into the classification process for the precise crop mapping, in order to reduce misclassification rates associated with low interclass and high intraclass spectral variability. Stochastic hidden Markov models (HMMs) are efficient yet computationally demanding classification approach which can simulate crop dynamics, exploiting the spectral information of their phenological states and the relations between these states. This letter aims to present a methodology which achieves accurate classification results while maintaining a low computational cost. A classification framework based on HMMs was developed, and different spectral indices were generated from the time series of Landsat ETM+ and RapidEye imagery, for modeling crop vegetation dynamics over a Mediterranean rural area, with high spatiotemporal crop heterogeneity. To further improve the HMMs indices classification, separability analysis and two different decision fusion strategies were tested. The assessment of the classification accuracy, along with an evaluation of the computational cost, indicated that the green-red vegetation index produced the most favorable results among the individual spectral indices. Although the decision fusion based on an integration of a reliability factor increased the overall accuracy by 3.1%, this came at the cost of computational time, compared to the separability analysis model which required less processing time.

Description: For hyperspectral imagery (HSI) classification, many works have shown the effectiveness of the spectral–spatial method. However, some previous works using neighboring information assumed that all neighboring pixels make an equal contribution to the central pixel, which is unreasonable for heterogeneous pixels, especially near the boundary of a region. In this letter, a nonlocal self-similarity based on the sparse coding method, followed by the use of a support vector machine classifier, is proposed to improve classification performance. Inspired by the success of nonlocal means, a new nonlocal weighted method is developed to determine the relationship between a test pixel and its neighboring ones. The nonlocal weights are determined by using the spectral angle mapper algorithm, which can exploit the spectral information of surface features. The experiments validate the superiority of our proposed method over existing approaches for HSI classification.

Description: Automatically detecting airports from remote sensing images has attracted significant attention due to its importance in both military and civilian fields. However, the diversity of illumination intensities and contextual information makes this task difficult. Moreover, auxiliary features both within and surrounding the regions of interest are usually ignored. To address these problems, we propose a novel method that uses a multiscale fusion feature to represent the complementary information of each region proposal, which is extracted by constructing a GoogleNet with a light feature module model that has an additional light fully connected layer. Then, the fusion feature is input to a support vector machine whose performance is enhanced using a hard negative mining method. Finally, a simplified localization method is applied to tackle the problem of box redundancy and to optimize the locations of airports. An experiment demonstrates that the fusion feature outperforms other features on airport detection tasks from remote sensing images containing complicated contextual information.

Description: This letter attempts to explore the potential sensitivity of the well-known spatial downscaling technique of coarse precipitation data to some bioclimatic stages of the Mediterranean area. For this purpose, first, an open data set covering a period of 15 years, including TRMM3B43, normalized difference vegetation index (NDVI), DEM, and rain gauge station measurements, was prepared. Then the NDVI-based spatial downscaling technique was applied over Morocco without taking account of bioclimatic stages. Second, based on the same data set, the key step of the downscaling approach (regression between TRMM3B43 and NDVI) was analyzed in five bioclimatic stages in order to assess the approach’s sensitivity. This letter demonstrated that the spatial downscaling approach performs well in the subhumid, semiarid, and in the arid bioclimatic stages, to a lesser extent. However, the approach seems to be sensitive and not adapted to the Saharan and humid stages.

Description: Manifold learning algorithms such as the isometric mapping (ISOMAP) algorithm have been widely used in the analysis of hyperspectral images (HSIs), for both visualization and dimension reduction. As advanced versions of the traditional linear projection techniques, the manifold learning algorithms find the low-dimensional feature representation by nonlinear mapping, which can better preserve the local structure of the original data and thus benefit the data analysis. However, the high computational complexity of the manifold learning algorithms hinders their application in HSI processing. Although there are a few parallel implementations of manifold learning approaches that are available in the remote sensing community, they have not been designed to accelerate the eigen-decomposition process, which is actually the most time-consuming part of the manifold learning algorithms. In this letter, as a case study, we discuss the graphics processing unit parallel implementation of the ISOMAP algorithm. In particular, we focus on the eigen-decomposition process and verify the applicability of the proposed method by validating the embedding vectors and the subsequent classification accuracies. The experimental results obtained on different HSI data sets show an excellent speedup performance and consistent classification accuracy compared with the serial implementation.

Description: During airborne transient electromagnetic (EM) surveys, transmitting and receiving antennas change their attitudes as the inevitable result of pilot maneuvers and natural forces, which makes the EM response different from that of the nominal attitudes when the antennas are straight and level. Attitude changes were usually neglected or partially considered in the past, which are not adequate for a quantitative interpretation. In this letter, we first scrutinize the mechanism of how the attitude change affects the EM response and divides these effects into two parts: the pure attitude effect and the resultant translation effect. Then, we introduce a novel method to involve the full attitude change in both modeling and inversion. Our compelling results finally demonstrate that the attitude change affects the early-time response much more than the late time and involving the full change in inversion can produce a better estimate of shallow geoelectric parameters.

Description: Determination of sea ice extent is important both for climate modeling and transportation planning. Detection and monitoring of ice are often done by synthetic aperture radar imagery, but mostly without any ground truth. For the latter purpose, robust and continuously operating sensors are required. We demonstrate that signals recorded by ground-based Global Navigation Satellite System (GNSS) receivers can detect coastal ice coverage on nearby water surfaces. Beside a description of the retrieval approach, we discuss why GNSS reflectometry is sensitive to the presence of sea ice. It is shown that during winter seasons with freezing periods, the GNSS-R analysis of data recorded with a coastal GNSS installation clearly shows the occurrence of ice in the bay where this installation is located. Thus, coastal GNSS installations could be promising sources of ground truth for sea ice extent measurements.

Description: A novel approach is presented to spatially disaggregate coarse soil moisture (SM) by only using remotely sensed vegetation index. The approach is based on the conditional relationship of vegetation with time-aggregated SM, allowing the coarse-scale SM to be disaggregated to the spatial resolution of the vegetation product. The method was applied to satellite-derived SM over January 2010–December 2011, using the high-resolution normalized difference vegetation index (NDVI). The results were evaluated against ground measurements during the two-year period over the contiguous United States and Spain, and also compared with an existing disaggregation method that also requires land surface temperature observations. It is shown that the proposed approach can provide fine-resolution SM with reasonable spatial variability.

Description: In radar high-resolution range profile (HRRP)-based target recognition, one of the most challenging tasks is the noncooperative target recognition with imbalanced training data set. This letter presents a novel recognition framework to deal with this problem. The framework is composed of two steps: first, the t-distributed stochastic neighbor embedding (t-SNE) and synthetic sampling are utilized for data preprocessing to provide a well segmented and balanced HRRP data set; second, a discriminant deep belief network (DDBN) is proposed to recognize HRRP data. Compared with the conventional recognition models, the proposed framework not only makes better use of data set inherent structure among HRRP samples for segmentation, but also utilizes high-level features for recognition. Moreover, the DDBN shares latent information of HRRP data globally, which can enhance the ability of modeling the aspect sectors with few HRRP data. The experiments illustrate the meaning of the t-SNE, and validate the effectiveness of the proposed recognition framework with imbalanced HRRP data.

Description: Many filtering algorithms have been developed to extract the digital terrain model (DTM) from dense urban light detection and ranging data or the high-resolution digital surface model (DSM), assuming a smooth variation of topographic relief. However, this assumption breaks for a middle-resolution DSM because of the diminished distinction between steep terrains and nonground points. This letter introduces a two-step semiglobal filtering (TSGF) workflow to separate those two components. The first SGF step uses the digital elevation model of the Shuttle Radar Topography Mission to obtain a flat-terrain mask for the input DSM; then, a segmentation-constrained SGF is used to remove the nonground points within the flat-terrain mask while maintaining the shape of the terrain. Experiments are conducted using DSMs generated from Chinese ZY3 satellite imageries, verified the effectiveness of the proposed method. Compared with the conventional progressive morphological filter method, the usage of flat-terrain mask reduced the average root-mean-square error of DTM from 9.76 to 4.03 m, which is further reduced to 2.42 m by the proposed TSGF method.

Description: The graph model-based semisupervised machine learning is well established. However, its computational complexity is still high in terms of the time consumption especially for large data. In this letter, we propose a fast semisupervised classification algorithm using the recently presented spatial-anchor graph for a large polarimetric synthetic aperture radar (Pol-SAR) data, named as Fast Spatial-Anchor Graph (FSAG) based algorithm. Based on an initial superpixel segmentation on the PolSAR image, the homogenous regions are obtained. The border pixels are reassigned to the most similar superpixel according to majority voting and distance measurement. Then, feature vectors are weighted within local homogenous regions. The refined spatial-anchor graph is constructed with these regions, and the semisupervised classification is conducted. Experimental results on synthesized and real PolSAR data indicate that the proposed FSAG greatly reduces time consumption and maintains the accuracy for terrain classifications compared with state-of-the-art graph-based approaches.

Description: Continuously expanding high-resolution and very high resolution multispectral image collections, provided by remote sensing satellites, require specific methods and techniques for data analysis and understanding. Even though there are several patch-based approaches for image classification and indexing, none of them are integrated within a standard. Having the goal to develop an MPEG-7 compliant descriptor for patch-based multispectral earth observation image classification and indexing, we propose a new feature extraction method able to extract maximum information from all the available spectral bands that Sentinel 2, the last generation of remote sensing satellites, provides. Using the polar coordinate transformation of the reflectance values, we obtain illumination invariant features, which can be used along with the scalable color descriptor present in MPEG-7 standard. Also, our method proves to enhance land cover classification of the areas affected by clouds and their shadows and provide similar classification results compared with the homogeneous texture descriptor (HTD), spectral histogram (SH), concatenated HTD with SH features, spectral indices (SIs), and bag-of-words-based descriptors, such as bag-of-SIs and bag-of-spectral-values on cloud-free areas.

Description: In this letter, we combine anisotropic and isotropic diffusion models and establish a combined energy variational model for seismic denoising. We propose a dynamic threshold to separate seismic sections into different feature areas and to choose different diffusion methods more precisely according to the characteristics of the seismic sections. Multilevel noise and multilevel edges can be treated automatically. Denoised results from a synthetic model and from field seismic sections demonstrate that our proposed model can suppress random noise and preserve the features of seismic sections efficiently.

Description: Using multiple types of features can effectively improve the classification accuracy of hyperspectral image (HSI). Multiple kernel learning (MKL) provides a flexible framework to fuse different features in a very natural way. In this letter, a novel MKL algorithm is proposed to integrate multiple types of features [i.e., principal components of original spectrum, multistructure morphological profiles (MPs), and multiattribute MPs] for HSI classification. The basic kernels are constructed with each feature subset separately, and the weights of basic kernels are determined by solving an optimization problem with the objective of the ideal kernel. Then, linear programming (LP) and signal sparse representation are adopted to solve the optimization problem, thus leading to two variants of the proposed algorithm, ideal kernel MKL (IKMKL)-LP and IKMKL-sparse, respectively. Experiments carried out on real hyperspectral data show that the proposed algorithms outperform several state-of-the-art MKL algorithms and reveal the capability of ranking the relevant features.

Description: The snow in high-mountain Asia (HMA) is of great importance, as it is very sensitive to the climate change. Air temperature and precipitation shifts/increases will be reflected in the timing of snowmelt onset. In this letter, a new algorithm is proposed to determine the snowmelt onset date from active and passive microwave remote sensing data, and the spatial and temporal pattern of snowmelt onset in HMA is studied using active and passive microwave remote sensing for the first time. Over 35 years of passive microwave data and ten years of active microwave data are used to derive the melt onset date in HMA. The active microwave data has 4.5-km resolution so that more detailed spatial pattern of snowmelt onset date can be derived compared to the 25-km resolution passive microwave data. Under climate change background, time series analyses of the snowmelt onset date in HMA are conducted to study the snowmelt onset time changes in recent 35 years. This letter provides an objective evidence of climate change impact on the cryospheric system. Time series analysis shows that the snowmelt onset date is becoming earlier in HMA region during 1988–2015, except the Karakorum Mountains and part of the western Kunlun Mountains. Mean air temperature is compared with the time series snowmelt onset date and the results show that there is strong correlation between mean air temperature and average snowmelt onset date. A 4.5 days/degree rate of snowmelt onset date advancing is found.

Description: The multitemporal interferometric synthetic aperture radar (InSAR) technique is a potential tool for measuring digital elevation models and surface deformation. It has the advantage of high precision, competitive spatial resolution, and wide coverage. To improve the accuracy of the final results, some adaptive multilook strategies have been proposed in which the identification of statistically homogeneous pixels (SHPs) is the key task. However, these methods are not always reliable in the case of small data sets. To improve this reliability, SHPs are identified based on the adaptive joint data vector comprising of temporal sample and spatial information in this letter. Additionally, the formulation of adaptive joint data vector is combined with local spatial features of SAR images. The presented adaptive multilook approach can be used in many interferometric applications, such as InSAR data filtering and coherence estimation. Experiments on six TerraSAR-X stripmap images of Tianjin in China validate the feasibility and effectiveness of the proposed approach.

Description: In mountainous areas, surface solar radiation (SSR) exhibits high spatiotemporal variation at different slopes and aspects due to its great topographic relief. To get mountain SSR spatial distribution, remote sensing-based methods have been popularly used, which separate SSR into direct solar radiation, diffuse sky radiation, and adjacent terrain radiation. However, the methods are highly depended on the atmospheric and angular information derived from different data sources. To clearly address the uncertainties associated with the estimation, this letter conducted a preliminary comparison study by using different atmospheric transmittance models and digital elevation model (DEM) data to retrieve SSR in the Mt. Gongga region. The comparison results indicated that the uncertainty of the atmospheric constituent data greatly limited the performance of the physical atmospheric transmittance models. The resolution of DEM data also played an important role in SSR determination because of the determination of surface angular information. High-resolution (30-m) DEM data showed better performance than low one (90 m). In addition, the systematic underestimation of SSR estimation with the empirical model was significantly improved by using the averaging method with nearby pixel values. It indicated that the geometric errors of satellite image and DEM data should be considered in the estimation.

Description: Multichannel singular spectrum analysis (MSSA) is an effective tool for processing multidimensional time-series such as the reconstruction of high-dimensional seismic data. Low-rank estimation is a key stage in MSSA algorithm that can affect the recovery process. Truncated singular value decomposition (TSVD) and singular value thresholding (SVT) are two conventional options for rank reduction, which, however, do not result in satisfactory outcomes, especially in low signal-to-noise-ratio cases. In this letter, we propose to leverage the optimal low-rank estimator that emerges from random matrix theory known as OptShrink. The OptShrink can obtain more robust low-rank estimation in comparison with TSVD and SVT. In addition, we propose to constrain the singular values using a damping factor. The proposed damped OptShrink method is applied on real and synthetic 3-D seismic data. The comprehensive experiments and discussion verify the superior reconstruction ability of the proposed method in comparison with MSSA.

Description: Recent advances have shown a great potential to explore collaborative representations in hyperspectral imagery (HSI) classification, including sparse representations and joint collaborative representations. In this letter, we propose a weighted regularized collaborative representation optimized classifier (WRCROC) that makes use of multiple collaborative representations. It strikes a balance between an optimized weighted joint collaborative representation classifier, which essentially classifies a test sample to the class that minimizes the distance between the sample and its representation in the selected class, and a weighted regularized collaborative representation classifier, which actually assigns a test sample to the class that minimizes the distance between the sample and its collaborative components. The proposed WRCROC algorithm is tested on two benchmark HSI data sets. Experimental results demonstrate that the proposed algorithm performs better than existing representation-based classifiers.

Description: This letter developes and validates a machine learning approach to forecast sea level anomalies (SLAs) derived from satellite altimetry in the tropical Pacific Ocean. The empirical orthogonal function (EOF), also known as principal component analysis, was used to extract dominant signals and reduce the dimensionality of data sets. Such dimensionality was decreased by describing spatial patterns (EOFs) and the corresponding temporal domains [principal components (PCs)]. Support vector regression (SVR) was employed to predict the time series obtained from the leading PCs. Thereafter, the temporal and spatial SLAs from the proposed EOFs were reconstructed to represent the spatiotemporal SLA prediction. Finally, the prediction result was compared with that of the conventional autoregressive integrated moving average (ARIMA) model. Both models reached satisfactory sea level predictions. Even so, intercomparison of the obtained results showed that the SVR significantly ( $P = 0.012$ ) outperformed the ARIMA model in sea level forecasting. That is, a considerably low root-mean-square error was attained for the differences between the predicted and observed mean SLAs.

Description: Existing global registration methods are prominently iterative. They require iterations and can be sensitive to point densities and noise. In contrast, closed-form solutions provide a more robust estimation model and do not involve iterations. In this letter, we present a global closed-form refinement for the terrestrial laser scanner (TLS) data registration problem. Our proposed method segments the task in three key steps. First, the method exploits a plane-based approach to compute the transformation parameters, resulting in the pairwise registration between point clouds. Second, we place all rotation parameters into a common coordinate system exploring one operation of the quaternions’ properties. Third, we constrain the refined rotation parameters to globally refine the translation. The effectiveness of the proposed method is demonstrated with a TLS data set. Experiments have demonstrated that the proposed method can properly create a consistent 3-D map of outdoor environments with accuracy at the decimeter level.

Description: We consider the tensor-based spectral-spatial feature extraction problem for hyperspectral image classification. First, a tensor framework based on circular convolution is proposed. Based on this framework, we extend the traditional principal component analysis (PCA) to its tensorial version tensor PCA (TPCA), which is applied to the spectral-spatial features of hyperspectral image data. The experiments show that the classification accuracy obtained using TPCA features is significantly higher than the accuracies obtained by its rivals.

Description: In recent years, transfer learning with pretrained convolutional networks (CNets) has been successfully applied to land-use classification with high spatial resolution (HSR) imagery. The commonly used transfer CNets partially use the feature descriptor part of the pretained CNets, and replace the classifier part of the pretrained CNets in the old task with a new one. This causes the separation and asynchrony between the feature descriptor part and the classifier part of the transferred CNets during the learning process, which reduces the effectiveness of the training process. To overcome this weakness, a transfer learning method with fully pretrained CNets is proposed in this letter for the land-use classification of HSR images. In the proposed method, a multilayer perceptron (MLP) classifier is quickly pretrained using the high-level features extracted by the feature descriptor of the pretrained CNets. Fully pretrained CNets can be generated by concatenating the feature descriptor of the pretrained CNets and the pretained MLP. Because both the feature descriptor and the classifier are pretrained, the separation and asynchrony between the two parts can be avoided during the training process. The final transferred CNets are then obtained by fine-tuning the fully pretrained CNets with the random cropping and mirroring strategy. The experiments show that the proposed method can accelerate the convergence of the training process with no loss of accuracy in land-use classification, and its performance is comparable to other latest methods.

Description: In this letter, a multifrequency experimental analysis is conducted for the estimation of the asphalt reflectivity and for the measurement of the radar cross section of some typical foreign object debris (FOD). The analysis is made with experimental data with a frequency between 10 and 77 GHz, acquired with a vector network analyzer and with-ingegneria dei sistemi 77-GHz radar prototype for FOD Detection. Experimental data acquired in a real operative scenario (runway of Taranto/Grottaglie airport) is also presented. The results show the possibility to detect an FOD target on an airport runway.

Description: Dielectric inversion of lunar permanently shadowed region (PSR) of moon poles has been studied for estimation of possible water–ice content. The Campbell model was directly applied to mini-SAR data for inversion on the Hermite-A crater region. However, this letter presents quantitative analysis that the lunar surface topography, i.e., surface roughness and slopes, and underlying dielectric media, and so on, can significantly affect this inversion. The model is actually degenerated into a half-space model without topographic account. This letter presents a two-layer model of Kirchhoff-approximation surface/small perturbation approximation subsurface to take account of all these topographic factors for PSR dielectric inversion.

Description: This letter aims to explore the potentialities of normalized difference water index (NDWI) and distance from sea to downscale coarse precipitation (TRMM3B43 product), whose contribution to downscaling precipitation remains unstudied. For this purpose, based on an open data set of 14 years, including TRMM3B43 and three predictors (NDWI, elevation, and distance from sea), stepwise regression and Akaike information criterion were applied in order to identify the best-fit models. The models that have given rise to best approximations and best-fits were used to downscale TRMM3B43 product, to a spatial resolution of 1 km. The resulting downscaled calibrated precipitations were validated by independent rain gauge stations (RGSs). The analysis exhibited that there is good and statistically significant correlations between TRMM3B43 and NDWI and a great agreement between downscaled precipitations and RGS measurements.

Description: Despite its ability to characterize 3-D environments, synthetic aperture radar (SAR) tomographic imaging, when applied to the characterization of targets concealed beneath forest canopies, may appear as an ill-conditioned estimation problem, with a complex mixture of numerous scattering mechanisms measured from a few different positions. Among the set of tomographic estimators that may be used to characterize such complex scattering environments, nonparametric tomographic techniques are more robust to focus on artifacts but limited in resolution and, hence, may fail to discriminate objects, whereas parametric ones provide better vertical resolution but cannot adequately handle continuously distributed volumetric scattering densities, characteristic of forest canopies. This letter addresses a new wavelet-based sparse tomographic estimation method for the 3-D imaging and discrimination of underfoliage objects that overcomes these limitations. The effectiveness of this new approach is demonstrated using L-band airborne tomographic SAR data acquired by the German Aerospace Center over Dornstetten, Germany.

Description: Object localization techniques have significant applications in civil fields and safety problems. A novel analytical formula is developed for accurate underwater and aerial object real-time localization by combining gravitational field and horizontal gravitational gradient anomalies. The proposed method enhances the accuracy of object localization and its excess mass estimation; it also effectively avoids the possible numerical instability and the singularity in the previous works. Finally, a synthetic underwater object navigation model was adopted to verify its performance. The results show that our newly developed method is more practical than existing methods.

Description: The Advanced Microwave Scanning Radiometer 2 (AMSR2) on board the Global Change Observation Mission-Water 1 launched in May 2012 provides brightness temperature at two different spatial resolutions globally with an average temporal resolution of two days. Surface soil moisture is retrieved using land parameter retrieval model (LPRM) and level-3 brightness temperature data (10.65- and 36.5-GHz channels). The present LPRM implementation has AMSR2 brightness temperatures only with dielectric and vegetation parameters derived from 10.65- and 36.5-GHz channels, respectively. Retrieved soil moisture in the Indian region for 0.25° and 0.1° grid cells is validated and compared over several sites with in situ measurements. It is observed that in comparing the LPRM retrieved soil moisture with field measurements over various sites, 0.1° grid performed relatively better over 0.25° grid.

Description: The ratio of averages is a robust edge detector which provides the property of constant false alarm rate for synthetic aperture radar (SAR) imagery. However, the rectangular window used in the calculation of local mean may cause numerous false maxima. The size of the processing window also has a significant effect on the detection performance, but it is difficult to determine the optimum window size. In this letter, we first propose a new ratio-based detector that is constructed by the Gabor odd filter. The scale of the proposed detector is related to the size of the processing window. Then, edge strength maps extracted by multiscale detectors are combined using an edge tracking algorithm to form a final response. We used the receiver operating characteristic curves to evaluate the performance of the proposed detector. The experimental results on simulated and real-world SAR images show that the proposed multiscale edge detector yields an accurate and consecutive edge response.

Description: In forest and agricultural scattering scenarios, the backscattered synthetic aperture radar (SAR) signature consists, depending on the frequency, of the superposition of ground and volume scattering contributions. Using multibaseline SAR data, SAR tomography techniques allow resolving contributions occurring at different heights. Two algorithms for the separation of ground and volume scattering are compared with respect to their ability to provide a coherent volume component that can be further used for parameter inversion, both of them requiring only the a priori known ground topography. Once the volume-only coherences are available, the total ground and volume scattering powers are estimated by means of a least squares fitting. The objective of this letter is to quantitatively evaluate the performance of this estimation by means of a Monte Carlo analysis with simulated data focusing on the impact of vertical resolution, errors in the knowledge of the ground topography and phase calibration residuals.

Description: This letter addresses an efficient algorithm for ground moving target detection and estimation of motion parameters by synthetic aperture radar (SAR). The proposed method outperforms the conventional robust principal component analysis (RPCA)–based ground moving target indication (GMTI) methods that were proposed in the literature. The ability to estimate the radial and along-track velocities of ground moving targets is provided. The rank constraint in the conventional RPCA problem will be automatically relaxed by employing a dictionary matrix for clutter representation. Thus, the new optimization problem will be solved easier with lower degrees of freedom. Furthermore, this dictionary helps to suppress the clutter of higher Doppler frequencies in case of wind blowing scenarios and intrinsic clutter motion modeling. By employing another dictionary matrix for all possible moving targets with different location and velocity, each solution of the optimization problem will be reasonable as it corresponds to a moving target. Although, the two dictionary matrices impose extra computational burden, this load will be prepared prior to other GMTI processing by the information of the SAR system and scenario parameters. Moreover, the algorithm is proposed for the single-channel SAR configurations and has lower computational load than the conventional RPCA-GMTI methods that have to process the recorded data of multichannel systems. Numerical and experimental results are used to evaluate the performance of the proposed method and validate the theoretical discussions.

Description: As is well known, multibaseline phase unwrapping (PU) is put forward to overcome single-baseline PU in discontinuous-terrain-height estimation. This letter presents a refined algorithm based on the cluster analysis (CA)-based noise-robust efficient multibaseline PU algorithm proposed by H. Yu. The basic idea is to combine multiple interferometric synthetic aperture radar interferograms with different baseline lengths by a linear combination. The new interferograms after the linear combination increase the ambiguity heights. The number of resulting groups on the envelope of the intercept histogram is decreased and the distance between different intercept groups is widened. Compared with the conventional CA method, the significant advantage of the refined CA (RCA) algorithm is that it improves noise robustness when the intercept groups are densely distributed. The proposed RCA algorithm is validated using the simulated interferometric data. The results demonstrate that the noise robustness performance is better than that of the CA method.

Description: An explicit expression for stochastic processes describing the dual-polarization weather radar echoes is presented. The probability distributions of the proposed model are defined in terms of the point values assumed by polarimetric and Doppler variables in the relevant radar sampling volume. The statistical properties of the model are discussed in order to verify its faithfulness as representative of real radar signals. The discussion considers the most general situation, i.e., it is not related to specific hydrometeor distributions or beam filling conditions.

Description: The Gaussian mixture model (GMM)-based methods have achieved great success in point set registration. However, they cannot be directly applied to image matching, because the features extracted from two images usually contain a large portion of outliers. In this letter, we propose a new method to extend the powerful GMM to the field of image feature points matching. The algorithm consists of two main steps. In the first step, points extracted from the images are mapped into a new subspace, in which feature similarity information is fused to get the new representation of the points. The second step performs an improved progressive process with the GMM to find correspondences satisfying the coherent constraint. In this way, finding correspondences among large outliers is feasible and the iteration converges faster. Experimental results on benchmark data sets show that the proposed method can find more correct matches with high accuracy.

Description: A proof-of-concept experiment was carried out to demonstrate the feasibility of retrieving snow water equivalent (SWE) using P-band signals of opportunity. The fundamental observation is the change in the phase of the reflected waveforms as related to the change in SWE. Through theoretical modeling it was found that the change in SWE was approximately linearly dependent on the change in phase. This was verified by retrieving SWE data collected and processed from a tower-based experiment at Fraser, CO, USA. A linear regression was performed on measured phase and in situ SWE. The correlation was found to be 0.94 and root mean square deviation was found to be 7.5 mm.

Description: In synthetic aperture radar (SAR) images, scattering centers (SCs) from the same geometric structure of the man-made target usually have the same scattering type and similar coordinates. Inspired by this observation, a novel clustering-based geometrical structure retrieval (C-GSR) method is proposed to estimate the geometrical structure of targets by clustering SCs according to their types and coordinates. The C-GSR method considers each peak in a SAR image as a single SC and extracts both frequency and polarization features for classification. Then, SCs are efficiently clustered using the density-distance-based clustering algorithm. Finally, the geometrical structure corresponding to each canonical scatterer can be retrieved by computing the coordinates of SCs associated with the corresponding cluster. Experimental results have demonstrated the feasibility and accuracy of the proposed C-GSR method.

Description: Horizon recognition is an important but difficult task for a seismic interpreter. A novel automatic recognition method based on the Kalman filter tracker is proposed. According to the properties of the seismic reflection profile, a pair of new linear state and measurement equations is built. Combined with three aided steps including the threshold detection, which finds the candidate measurements, the logic method, which automatically starts a new possible horizon, and the probabilistic data association, which judges the probability of each candidate measurement belonging to each of all the existing horizons, the Kalman filter tracker can reduce the noise involved in the candidate measurements and recognize the multihorizons automatically and simultaneously. Unlike the common picking methods, the proposed method can get the detailed time-space location data for each identified horizon, which is very important for the followed processing and interpretation. The experimental results on the synthetic and real data show that the proposed method can not only track the faint horizons, but also improve the continuity of horizons. The proposed method outperforms the common Canny edge detector.

Description: This letter presents an extension of the free-space method for the dielectric characterization of Eucalypt litter fire using the measured transmission phase shift of a propagating signal over a frequency range of 5–40 GHz. This method is especially suitable for a broadband, routine, nonintrusive, and accurate evaluation of fire dielectric properties from the measured transmission coefficient ( ${S}_{21}$ ). A 72 cm $times $ 47 cm rectangular brick burning area prepared with Eucalypt ground litter was equipped with four K-type thermocouples at different heights to measure flame temperatures, which ranged from 283–865 K. The measured phase shift of an electromagnetic signal transmitted through the Eucalypt fire varied from 12.2 to −166.7° at 5.35 and 39.74 GHz. From this phase shift, the calculated real part of the relative permittivity of the fire was between 0.877 and 1, with the imaginary part from 0.02 to 0.09. The relative permittivity of less than unity confirms the formation of plasma in the combustion zone of fire, and shows good agreement with previous research. However, in this letter, the combustion zone region is defined with significantly higher accuracy, and characteristics are evaluated over a much broader frequency range. The extracted parameters are important for understanding of the reaction of electromagnetic waves in wildfire environments and can be used to design radar systems for forest fire detection.

Description: In this letter, a novel estimation approach for the Doppler chirp rate and baseline in azimuth is proposed for the multichannel in azimuth high-resolution and wide-swath (HRWS) synthetic aperture radar (SAR) system. First, a range-invariant Doppler chirp rate estimation approach is developed based on a map drift algorithm and correlation function method. Then, a weighted local maximum-likelihood approach is adopted to obtain an accurate estimation of the range-variant Doppler chirp rate. With an accurate Doppler chirp rate, the baseline in azimuth can be estimated, which is developed from the correlation function between the echoes of adjective channels. The proposed approaches are successfully applied to process real five-channel HRWS SAR echo data, demonstrating the efficacy of the proposed methods.

Description: This letter presents a new method to generate superpixel lattice. Unlike the existing methods based on strip seaming, it iteratively refines an initial lattice to achieve an expected lattice. In each iteration, it adjusts local lattice step by step. At each step, a vertex and its four connecting edges are refined by finding the optimal horizontal and vertical edges. The optimal edges are found by dynamic programming technique. This method is tested against both remotely sensed images and standard data set dedicated to segmentation evaluation. The proposed method is better or competitive with the existing state-of-the-art superpixel lattice methods concerning both performance and efficiency.

Description: We present a solution for simplifying a recently proposed two-step processing technique that allows to retrieve the constant phase offset present in the unwrapped synthetic aperture radar (SAR) interferograms by exploiting an external, even low-accuracy, digital elevation model (DEM) of the illuminated area and without using corner reflectors. In particular, we show in this letter that the second processing step, namely, the slope-topography-based estimate, can be avoided without impairing the accuracy of the final phase offset estimate. To this aim, we introduce a simple modification to the first step, referred to as phase-based estimate, by considering the vertical bias of the available external DEM as the second unknown parameter in the carried out estimation. The simplified algorithm is very easy to implement and is particularly suitable for airborne SAR interferometry. It has been tested on real airborne SAR data and the obtained results show that the achieved accuracy is the same or better than that achieved through the original two-step approach.

Description: Over the last few years, several new strategies for spectral unmixing of remotely sensed hyperspectral data have been proposed. Many of them have been developed to solve the most time-consuming and relevant step: endmember extraction. However, unmixing algorithms can be computationally very expensive in terms of processing time and energy consumption, a fact that compromises their use in applications under real-time and energy/power constraints. In this letter, we present a new parallel simplex growing algorithm (SGA) for hyperspectral data which exploits the memory hierarchy with operations in single-precision floating point. Those optimizations accelerate the most time-consuming parts of this method using the open computing language (OpenCL) standard. We have evaluated the performance versus energy consumption using the same open standard for parallel programming over a diverse set of heterogeneous platforms. Experiments have been conducted using real hyperspectral images collected by NASA’s Airborne Visible Infrared Imaging Spectrometer and a collection of 24 synthetic hyperspectral images simulated with different sizes and number of endmembers (10–30). Considering the power consumption and OpenCL across all the proposed devices, the analysis presented indicates that the SGA can now be executed in computationally efficient fashion, which was not possible before introducing the parallel implementation described in this letter.

Description: Vegetation water content (VWC) is an important parameter of agriculture and forestry. In this letter, specific polarization ratios were evaluated for estimating VWC of corn and soybean. Backscattering coefficients ( $mathrm {boldsymbol{sigma }}_{mathbf {hh}}, mathrm {boldsymbol{sigma }}_{mathbf {vv}}$ , $mathrm {boldsymbol{sigma }}_{mathbf {vh}}$ and $mathrm {boldsymbol{sigma }}_{mathbf {hv}}$ ), polarization ratios ( $mathrm {boldsymbol{sigma }}_{mathbf {hh}}mathrm {mathbf {/}}mathrm {boldsymbol{sigma }}_{mathbf {vv}}$ , $mathrm {boldsymbol{sigma }}_{mathbf {vv}}mathrm {mathbf {/}}mathrm {boldsymbol{sigma }}_{mathbf {vh}} $ , and $mathrm {boldsymbol{sigma }}_{mathbf {hh}}mathrm {mathbf {/}}mathrm {boldsymbol{sigma }}_{mathbf {hv}}$ ), and the radar vegetation index derived from L-band (1.26 GHz) and S-band (3.15 GHz) radar data of the passive and active L- and S-band sensor (PALS) in Soil Moisture Experiments 2002 were implemented to develop various linear relationship models with field VWC measurements for corn and soybean, respectively. L-band $mathrm {boldsymbol{sigma }}_{mathbf {hh}}mathrm {mathbf {/}}mathrm {boldsymbol{sigma }}_{mathbf {vv}}$ was found to be most correlated with corn VWC (R = 0.81), while for soybean, L-band $mathrm {boldsymbol{sigma }}_{mathbf {hh}}mathrm {mathbf {/}}mathrm {boldsymbol{sigma }}_{mathbf {hv}}$ - /inline-formula〉 was the best parameter to estimate VWC with an R of 0.90. Based upon these analyses, prediction equations for the estimation of corn and soybean VWC using the polarization ratios were developed. Results indicated that L-band $mathrm {boldsymbol{sigma }}_{mathbf {hh}}mathrm {mathbf {/}}mathrm {boldsymbol{sigma }}_{mathbf {vv}}$ was able to estimate corn VWC with a root mean square error (RMSE) of 0.53 kg/m 2 and a mean absolute relative error (MARE) of 11.48%. As for soybean, L-band $mathrm {boldsymbol{sigma }}_{mathbf {hh}}mathrm {mathbf {/}}mathrm {boldsymbol{sigma }}_{mathbf {hv}}$ was capable of estimating soybean VWC with an RMSE of 0.12 kg/m 2 and an MARE of 13.33%. The main reason for these differences is most likely due to the disparate structure features and VWC distribution of corn and soybean. This letter proposes an effective method for acquiring VWC in regional areas, and it is also considered to be a powerful supplement for the current methods based on optical remotely sensed data.

Description: Remote sensing of top-of-canopy (TOC) long-term sun-induced chlorophyll fluorescence (SIF) is necessary to better understand the SIF-photosynthesis relationship. Statistical methods provide an alternative to TOC SIF retrieval, as they are independent of synchronous irradiance measurements and may better describe actual irradiance. This letter aims to evaluate the feasibility of using statistical methods for time series TOC SIF retrieval in the absence of synchronous irradiance measurements. Results show that the training set should include nonfluorescent radiance spectra under a variety of solar zenith angles, and that water vapor is an important contributor of spectral variation within 717–745 nm. On the diurnal scale, atmospheric features trained from irradiance spectra can be used to retrieve SIF values from high-frequency upwelling radiance spectra. Features independently trained from nonfluorescent radiance spectra measured on one day can be used for SIF retrieval on a different day within a relatively short period. Our results show that statistical methods have the potential to simplify ground-based SIF measurements and data processing.

Description: Most applications of frequency-modulated continuous-wave radar described in the literature involve targets that are in relatively close proximity to the radar. In these cases, the round-trip travel time of the target’s radar signature is small relative to the transmit chirp duration, simplifying the processing required for range and velocity extraction. This is not the case for more distant targets, where much of the radar signature is received after the start of the subsequent transmit waveform. In this letter, we examine various signal-processing options for coping with this long-range condition. We analytically demonstrate how to retain both range and Doppler shift information for an arbitrary number of targets spaced anywhere from very near the target up to the radar’s unambiguous range. The motivation for this work is to develop a 95-GHz Doppler radar for measuring ice and dust particle dynamics in cometary jets. Simulations and experimental results are provided to validate our methods.

Description: The existence of noise in hyperspectral ima-gery (HSI) seriously affects image quality. Noise removal is one of the most important and challenging tasks to complete before hyperspectral information extraction. Though many advances have been made in alleviating the effect of noise, problems, including a high correlation among bands and predefined structure of noise covariance, still prevent us from the effective implementation of hyperspectral denoising. In this letter, a new algorithm named the penalized linear discriminant analysis (PLDA) and noise adjusted principal components transformation (NAPCT) was proposed. PLDA was applied to search for the best noise covariance structure, while the NAPCT was employed to remove the noise. The results of the tests with both HJ-1A HSI and EO-1 Hyperion showed that the proposed PLDA-NAPCT method could remove the noise effectively and that it could preserve the spectral fidelity of the restored hyperspectral images. Specifically, the recovered spectral curves using the proposed method are visually more similar to the original image compared with the control methods; quantitative matrices, including the noise reduction ration and mean relative deviation, also showed that the PLDA-NAPCT produced less bias than the control methods. Furthermore, the PLDA-NAPCT method is sensor-independent, and it could be easily adapted for removing the noise from different sensors.

Description: We present an approach to segment vehicle tracks in coherent change detection images, a product of combining two synthetic aperture radar images taken at different times. The approach uses multiscale higher order random field models to capture track statistics, such as curvatures and their parallel nature, that are not currently utilized in existing methods. These statistics are encoded as 3-by-3 patterns at different scales. The model can complete disconnected tracks often caused by sensor noise and various environmental effects. Coupling the model with a simple classifier, our approach is effective at segmenting salient tracks. We improve the F-measure on a standard vehicle track data set to 0.963, up from 0.897 obtained by the current state-of-the-art method.

Description: A ground-wave technique is introduced in this letter to directly extract the pavement permittivity and thickness from measured data of ground-coupled ground-penetrating radar (GPR). Analytic solution, numerical simulation, and experimental test are carried out to validate the method. This technique enables bistatic radar to obtain both thickness and permittivity by just one measurement, which effectively reduces measurement and computation time for GPR applications.

Description: This letter addresses the problem of ionospheric scintillation effects on the global navigation satellite system (GNSS) signals. Severe scintillations degrade the signal intensity below the fade margin of the GNSS receiver, resulting in failure of the positioning and navigational services. A robust methodology is needed for the estimation and mitigation of such ionospheric scintillation effects. Hence, in this letter, the application of an adaptive signal decomposition technique based on variational mode decomposition (VMD), in combination with the detrended fluctuation analysis (DFA) method, is reported. VMD-DFA effectively decomposes the GNSS signal affected by ionospheric scintillations into a number of intrinsic mode functions and provides a threshold for the detection and mitigation of scintillations noise. Monte Carlo simulation results demonstrate that the proposed algorithm is superior and reliable for eliminating the amplitude scintillation effects compared to the complementary ensemble empirical mode decomposition method. The application of the proposed algorithm on both synthetic (Cornell scintillation model) and real-time measured GNSS data obtained from GNSS software navigation receiver at Rio de Janeiro, Brazil, has shown its potentiality in mitigating the ionospheric amplitude scintillation effects.

Description: Predictive coding is attractive for compression of hyperspectral images onboard of spacecrafts in light of the excellent rate-distortion performance and low complexity of recent schemes. In this letter, we propose a rate control algorithm and integrate it in a lossy extension to the CCSDS-123 lossless compression recommendation. The proposed rate algorithm overhauls our previous scheme by being orders of magnitude faster and simpler to implement, while still providing the same accuracy in terms of output rate and comparable or better image quality.

Description: In this letter, we propose a novel method for unsupervised change detection in multitemporal multispectral Landsat images using multiobjective evolutionary algorithm (MOEA). The proposed method minimizes two different objective functions using MOEA to provide tradeoff between each other. The objective functions are used for evaluating changed and unchanged regions of the difference image separately. The difference image is obtained by using the structural similarity index measure method, which provides combination of the comparisons of luminance, contrast, and structure between two images. By evolving a population of solutions in the MOEA, a set of Pareto optimal solution is estimated in a single run. To find the best solution, a Markov random field fusion approach is used. Experiments on semisynthetic and real-world data sets show the efficiency and effectiveness of the proposed method.

Description: With the continuous improvement of image resolution, details on aerial images provide abundant available information for vehicle detection. Nevertheless, traditional works mainly exploited the overall information of the vehicles ignoring the local details, such as front and rear windshields, and thus, there were usually more than 15% false alarms in the final vehicle detection results. In this letter, we propose a vehicle detection method making full use of high level details on aerial images. In the training stage, we choose front windshield samples to train a part detector and whole vehicle samples to train a root detector. In the matching stage, we first use the root detector to define an entire vehicle obtaining the root response, then the part detector is scanned in the root bounding box to decide a front windshield and get the part response. Afterward, the part response is transformed by setting weight w based on the part position offset. More importantly, contextual information is appropriately used in the process of determining the part position offset. Final detection score is the combination of root response and the transformed part response. We have demonstrated that the proposed method has achieved better performance with more than 6.43% increase of correct detection rate and more than 5.63% decrease of false detection rate compared with the state-of-the-art approaches.

Description: This letter proposes a method using convolutional neural networks (CNNs) for airport detection on optical satellite images. To efficiently build a deep CNN with limited satellite image samples, a transfer learning approach had been employed by sharing the common image features of the natural images. To decrease the computing cost, an efficient region proposal method had been proposed based on the prior knowledge of the line segments distribution in an airport. The transfer learning ability on deep CNN for airport detection on satellite images had been first evaluated in this letter. The proposed method was tested on an image data set, including 170 different airports and 30 nonairports. The detection rate could reach 88.8% in experiments with seconds’ computation time, which showed a great improvement over other the state-of-the-art methods.

Description: In the summer of 2016, severe storms caused serious casualties and destruction of facilities and properties over South China. Near-real-time (NRT) satellite precipitation products are attractive to rainstorm monitoring and flood warning guidance owing to its combination of timeliness, high spatiotemporal resolution, and broad coverage. We evaluate the performance of four NRT satellite products, i.e., Precipitation Estimation from Remotely Sensed Information using Artificial Neural Networks, 3B42RT, Global Satellite Mapping of Precipitation (GSMaP) NRT, and Integrated Multi-satellitE Retrievals for Global Precipitation Measurement (IMERG) Late run using a high-quality merged product in the rainy June over South China. In addition, a method based on an empirical flash flood guidance and the Flash Flood Potential Index is proposed to examine the applicability of satellite products in guiding flood warning. The IMERG Late run and GSMaP NRT perform the closest-to-ground observations. 3B42RT detects the most flood warning events due to its notable overestimation of actual precipitation. We recommend that the IMERG Late run is the best NRT satellite product in capturing flood hazard events according to the Pareto Efficiency of jointly optimizing higher hit ratio and lower false alarms.

Description: This letter aims to estimate the effective scattering albedo ( $\omega _{p}$ ) over the tropical forests using L-band (1.4 GHz) microwave remote sensing. It is carried out using Soil Moisture and Ocean Salinity (SMOS) mission data over five years (2011–2015). We find similar values of $\omega _{p}$ computed over the Congo and Amazon forests. The $\omega _{p }$ values depend slightly on the polarization. The values of $\omega _{p }$ at H-polarization and at 52° ± 5° (40° ± 5°) of incidence angle are within the range 0.064 – 0.069 ± 0.01 (0.061 – 0.067 ± 0.012). At V-polarization, the values of $\omega _{p }$ are slightly lower (0.060 – 0.061 ± 0.013 at 52° ± 5° of incidence angle and 0.052 – 0.055 ± 0.013 at 40° ± 5° of incidence angle). These findings should contribute to a better calibration of the value of $\omega _{p }$ over the tropical forests in both the SMOS and SM active and passive retrieval algorithms, leading to increase the SM retrieval accuracy over heterogeneous pixels.

Description: Optimal stacking of multiple data sets plays a significant role in many scientific domains. The quality of stacking will affect the signal-to-noise ratio and amplitude fidelity of the stacked image. In seismic data processing, the similarity-weighted stacking makes use of the local similarity between each trace and a reference trace as the weight to stack the flattened prestack seismic data after normal moveout correction. The traditional reference trace is an approximated zero-offset trace that is calculated from a direct arithmetic mean of the data matrix along the spatial direction. However, in the case that the data matrix contains abnormal misaligned trace, erratic, and non-Gaussian random noise, the accuracy of the approximated zero-offset trace would be greatly affected, and thereby further influence the quality of stacking. We propose a novel weighted stacking method that is based on principal component analysis. The principal components of the data matrix, namely, the useful signals, are extracted based on a low-rank decomposition method by solving an optimization problem with a low-rank constraint. The optimization problem is solved via a common singular value decomposition algorithm. The low-rank decomposition of the data matrix will alleviate the influence of abnormal trace, erratic, and non-Gaussian random noise, and thus will be more robust than the traditional alternatives. We use both synthetic and field data examples to show the successful performance of the proposed approach.

Description: For real-time and high-resolution radiometric imaging, the synthetic aperture interferometric radiometer (SAIR) technique seems to solve the tradeoff between imaging resolution and aperture size. Indeed, it can synthesize a large antenna array by sparsely arranging a small number of antennas to achieve high spatial resolution. Nevertheless, a conventional interferometric radiometer requires as many receivers as antennas needed. This constraint is one of the limitations of SAIR regarding the hardware cost, the system complexity, and the computation load. In this letter, a compressed acquisition technique is proposed to collect the antenna signals. In this method, the antenna’s signals are coded and combined red with a passive microwave device. From the measured waveforms at the receiver’s outputs, a decoding process is performed to estimate the received signal by each antenna. This allows the computation of the visibility function required for image reconstruction. The effectiveness of the proposed method has been tested by means of simulations analysis and has experimentally been applied to a punctual noise source detection. The results reveal that this method can be an efficient alternative for the simplification of the conventional interferometric radiometers’ architectures.

Description: Despite its ability to handle occlusions and noise, sparse tracking may be inadequate to describe complex noise corruption, for instance, in urban road tracking, where road surfaces are often significantly disrupted by the existence of occlusions and noise in high-resolution (HR) satellite imagery. To address this issue, this letter presents a semiautomatic approach for road extraction from HR satellite images. Firstly, a multifeature sparse model is introduced to represent the road target appearance. Next, a novel sparse constraint regularized mean-shift algorithm is used to support the road tracking. Furthermore, multiple features are combined by weighting their contributions using a novel reliability measure derived to distinguish target from background. The experiments confirm that the proposed method performs better than the current state-of-the-art methods for the extraction of roads from HR imagery, in terms of reliability, robustness, and accuracy.

Description: Raw signal simulators for synthetic aperture radar (SAR) in different acquisition modes have been studied individually. This letter is dedicated to presenting a method for simulating multimode SAR raw signal in a unified framework. To this end, we first simulate stripmap SAR raw signal, and then mutate the raw signal from stripmap mode into the desired acquisition mode. The acquisition mode mutation is implemented by azimuth-time-variant and range-frequency-dependent bandpass filtering. While processing in range-frequency domain brings forth high accuracy, processing in azimuth-time domain makes the method applicable for flexible antenna steering laws specified by multiple SAR acquisition modes, such as staring spotlight mode, sliding spotlight mode, and terrain observation by progressive scans mode. The proposed method is validated by the simulation results.

Description: It is well known that both gravity and magnetolluric (MT) methods can be used for the depth-to-basement estimation due to the density and conductivity contrast between the sedimentary basin and the underlaid basement rocks. In this case, the primary targets for both methods are the interface between the basement and sedimenary rocks as well as the physical properties of the rocks (density and conductivity). The solution of this inverse problem is typically nonunique and unstable, especially for gravity inversion. In order to overcome this difficulty and provide a more robust solution, we have developed a method of joint inversion to recover both the depth to the basement and the physical properties of the sediments and basement using gravity and MT data simultaneously. The joint inversion algorithm is based on the regularized conjugate gradient method. To speed up the inversion, we use an effective forward modeling method based on the surface Cauchy-type integrals for the gravity field and the surface integral equation representations for the MT field, respectively. We demonstrate the effectiveness of the developed method using several realistic model studies.

Description: In this letter, we propose an imaging algorithm for the holographic synthetic aperture radar tomography in the circumstance of sparse and nonuniform elevation circular passes. Considering the anisotropic behavior of scatterers and the off-grid effect of sparse signal recovery, the algorithm combines the 2-D adaptive imaging method for circular SAR and the sparse Bayesian inference-based method for elevation reconstruction. For each circular pass, the azimuth-range 2-D image can be formed by the adaptive imaging method, which depends on the preretrieved maximum azimuth response angle and the azimuth persistence width. To deal with the off-grid effect in elevation reconstruction, which is caused by the deviation between the true scatterers and the discretized imaging grids, the off-grid sparse Bayesian inference method jointly estimates the scatterers and elevation off-grid error by applying their hierarchical priors. Compared with the conventional compressive sensing method that does not concern the off-grid effect, the proposed algorithm can provide more accurate 3-D reconstruction for pointlike targets, which is verified by the real-data experiments.

Description: Forests are simplified as homogeneous volumes constituted of randomly uniform particles, characterized by a constant extinction coefficient in the random volume over ground (RVoG) model, which has been extensively applied to polarimetric synthetic aperture radar (SAR) interferometry for forest height estimation. This letter takes into account the heterogeneous vertical structure reflected by the vertically varying extinction coefficient curve in the forest volume layer, and modifies the RVoG model to make it be more suitable for height inversion of forests with complex structures. For this purpose, the normalized extinction coefficient curve is fit by large-footprint light detection and ranging full waveform data using the Gaussian function. Finally, the varying extinction RVoG model is applied to forest height estimation using airborne L-band SAR data acquired by the E-SAR system. The results are compared with in situ measurements, which indicate that the varying extinction RVoG model can obtain more accurate results for forest height inversion.

Description: Recent light detection and ranging (lidar) systems using photon-counting technology are able to collect data with significantly higher efficiency compared with the current commercially available linear-mode lidar systems. However, the high quantum sensitivity of single-photon lidar (SPL) systems results in noisy point clouds due to the influence of solar noise and dark count returns. Therefore, an effective noise removal algorithm is required to interpret SPL data. The uneven distribution of noise returns and the removal of noise close to signal returns are two significant challenges for SPL filtering. In this letter, a novel adaptive ellipsoid searching (AES) method is proposed. The AES uses a spherical noise density estimation model and a morphing ellipsoid determined by local principal components. The proposed method was tested on Sigma Space high-resolution quantum lidar system (HRQLS) SPL data sets and the results were compared with voxel-based filtering of the same data. Independent comparisons of each filtered result with coincident linear-mode airborne lidar data were also undertaken. We find that the root mean square error of the AES results on solid planes is 0.09 versus 0.11 m for voxel-based, 0.12 versus 0.14 m for bare ground, and 2.07 versus 2.55 m for vegetation canopy. We also used manually selected solid planar surfaces as a reference and find that the proposed method successfully removed twice as many noise points as the voxel-based method.

Description: P-wave time picking is of great significance in microseismic data processing. However, traditional time picking methods do not consider the difference of low-dimensional manifold features between signal and noise which can be extracted more effectively in low signal to noise ratio scenarios. In this letter, we develop a new method named spectral multimanifold clustering for picking P-wave arrivals. It can extract the low-dimensional manifold features from a suitable affinity matrix. In this approach, the manifold features of data are concentrated by residual statics estimation and a suitable affinity matrix is constructed using structural similarity and local similarity. Then, by using unnormalized spectral clustering, the low-dimensional manifold features extracted from the affinity matrix can be classified into noise cluster and signal cluster. Finally, the initial time of the signal cluster is considered to be the first arrival time in microseismic data. We design a series of experiments using both synthetic and field microseismic data. Our proposed method demonstrates higher accuracy, better stability, and noise immunity than either the short and long time average method or the akaike information criterion method.

Description: Super-resolution is an image processing technology that recovers a high-resolution image from a single or sequential low-resolution images. Recently deep convolutional neural networks (CNNs) have made a huge breakthrough in many tasks including super-resolution. In this letter, we propose a new single-image super-resolution algorithm named local–global combined networks (LGCNet) for remote sensing images based on the deep CNNs. Our LGCNet is elaborately designed with its “multifork” structure to learn multilevel representations of remote sensing images including both local details and global environmental priors. Experimental results on a public remote sensing data set (UC Merced) demonstrate an overall improvement of both accuracy and visual performance over several state-of-the-art algorithms.

Description: Inpainting techniques based on partial differential equations (PDEs), such as diffusion processes, are gaining growing importance as a novel family of image compression methods. Nevertheless, the application of inpainting in the field of hyperspectral imagery has been mainly focused on filling in missing information or dead pixels due to sensor failures. In this letter, we propose a novel PDE-based inpainting algorithm to compress hyperspectral images. The method inpaints separately the known data in the spatial and spectral dimensions. Then, it applies a prediction model to the final inpainting solution to obtain a representation much closer to the original image. Experimental results over a set of hyperspectral images indicate that the proposed algorithm can perform better than a recent proposed extension to prediction-based standard CCSDS-123.0 at low bit rate, better than JPEG 2000 Part 2 with the DWT 9/7 as a spectral transform at all bit rates, and competitive to JPEG 2000 with principal component analysis, the optimal spectral decorrelation transform for Gaussian sources.

Description: Shipborne high-frequency surface wave radar (SHFSWR) has exhibited great advantages over onshore HFSWR (OHFSWR) in ocean remote sensing. Unlike OHFSWR, SHFSWR suffers the problem of Doppler spectrum spread owing to platform movement, which is a great challenge preventing the extraction of ocean surface parameters for SHFSWR. To address this challenge, in this letter, the mathematical model of ocean surface wind direction is first investigated based on the first-order SHFSWR cross section. Furthermore, a method for the wind direction inversion without ambiguity from the spread Doppler spectrum is proposed using a single receiving antenna. Meanwhile, the wind directions of the sea area covered by radar can be obtained by sequentially utilizing the proposed method, which is more appropriate for the application of SHFSWR with limited deck space and less cost. Experimental results of the real data collected in Taiwan Strait preliminarily verify the detection accuracy and the distance limit of the wind direction inversion, as the root-mean-square error and the detection range are 9.85° and 120 km, respectively

Description: Many synthetic aperture radar (SAR) image exploitation tasks rely on the presence of certain features in the SAR images to be investigated. SAR simulation is an important tool to study such signatures for a wide range of aspect and incidence angles, for which usually real data cannot be provided. Furthermore, SAR simulation offers the opportunity to change the 3-D model so that many features can be traced back directly to features in the underlying 3-D scene. In this letter, the second approach is used to investigate the SAR signature of oil tanks with fixed and floating roofs. The knowledge gained by SAR simulation is then used to estimate the height of both types of tanks and the filling height of floating roof tanks.

Description: In this letter, we propose a new despeckling method based on two-step sparse decomposition. First, the grouping by block matching method identifies similar image patches and stacks them into a group, so that the group of similar patches are mostly homogeneous, which is suitable for the followed sparse decomposition method. And then, the proposed two-step sparse decompositions are applied to each group. The first sparse decomposition is a classical sparse representation to obtain an overcomplete dictionary and the sparse coefficients. The second sparse decomposition is a subspace decomposition over the dictionary. We proposed a measurement from the sparse coefficients as the criterion to identify a principal signal subdictionary. Finally, the image is reconstructed by the linear combination of the atoms of the principal subdictionary. The proposed method takes benefits from learned overcomplete dictionary, which fully explores details and from the principal subdictionary, which reduces strong noises. Experimental results demonstrate the efficiency of the proposed method to denoise synthetic aperture radar images. Our method can achieve high performances in terms of both structure details preservation and speckle noise reduction.

Description: The vast amount of data obtained from current remote sensing data acquisition technologies represents a wealth of useful and affordable geospatial data for policy and decision makers. However, the consequent computational cost of analyzing these data may become prohibitive. This letter extends previous efforts in exploiting distributed processing to speed up the image interpretation process. In this letter, we propose and evaluate a mechanism to exploit task parallelism in addition to data parallelism. Experiments conducted on cloud computing infrastructure, following an object-based interpretation model, demonstrated that substantial performance gains can be obtained with the proposed mechanism.

Description: The conventional method of extracting ocean surface currents by high-frequency over-the-horizon radar is based on the fixed first-order Bragg frequency formula and ignores the effects caused by the environment, especially in near-shore areas. In this letter, a current inversion model based on 2-D Fourier series expansion was developed. The first-order Bragg frequency and the Doppler offset induced by radial current are dealt with the bivariate functions of group distance and azimuth angle in the proposed method. By solving an overdetermined matrix equation with the least-square fitting method, the current at each detection grid can be estimated. As the Bragg frequency obtained by this new algorithm is adaptive to the environment, the accuracy of current measurement will be improved. The feasibility and effectiveness of the new method are verified with simulations and experimental results. The currents estimated by the traditional method and the new algorithm are compared with two in situ buoys. Results indicate that the new algorithm possesses comparable accuracy for far-shore areas and better accuracy for near-shore areas when compared with the conventional method.