IEEE Transactions on Geoscience and Remote Sensing

Syndicate content
TOC Alert for Publication# 36
Updated: 2 days 19 hours ago

Front Cover

Wed, 05/01/2019 - 00:00
Presents the front cover for this issue of the publication.

IEEE Transactions on Geoscience and Remote Sensing publication information

Wed, 05/01/2019 - 00:00
Presents a listing of the editorial board, board of governors, current staff, committee members, and/or society editors for this issue of the publication.

Table of contents

Wed, 05/01/2019 - 00:00
Presents the table of contents for this issue of the publication.

A Long-Term Historical Aerosol Optical Depth Data Record (1982–2011) Over China From AVHRR

Wed, 05/01/2019 - 00:00
A long-term historical aerosol optical depth (AOD) data set from 1982 to 2011 over China (15-45° N; 75-135° E) with 0.1 spatial resolution has been produced from Advanced Very High Resolution Radiometer (AVHRR) Pathfinder Atmospheres-Extended level-2B data. The spatial distribution pattern shows that high AOD values are found in central and eastern China over the entire period with AODs larger in summer and spring than in autumn and winter. As the high-quality products from AERONET were absent for this period over mainland China, AOD data obtained using the broadband extinction method from solar radiation stations have been used to verify the quality of the AVHRR AOD data set over China. The intercomparison results show that the interannual variation of AOD has been well captured in the variation curve of the AOD monthly mean and the variation trend is also consistent over the whole period. The correlation coefficient of the monthly mean is mostly larger than 0.55, the agreement index is larger than 0.57, and the relative error is less than 21%. Both AVHRR and visibility data sets show high values in regions with rapid economic development. Using Moderate Resolution Imaging Spectroradiometer AOD data as references, it is found that AVHRR AOD from this paper has better accuracy in general than that from Deep Blue (DB) algorithm over China, especially over eastern and southern China, while DB provides more coverage especially over bright surface such as northwest China. This long-term historic AOD data set can be used together with other AOD data sets to study the climate and environmental changes, especially in the 1980s and 1990s.

Monitoring Surface Phenomena Created by an Underground Chemical Explosion Using Fully Polarimetric VideoSAR

Wed, 05/01/2019 - 00:00
Sandia National Laboratories flew its Facility for Advanced RF and Algorithm Development X-Band (9.6-GHz center frequency), fully polarimetric synthetic aperture radar (PolSAR) in VideoSAR mode to collect complex-valued SAR imagery before, during, and after the sixth Source Physics Experiment's (SPE-6) underground explosion. The VideoSAR products generated from the data sets include “movies” of single- and quad-polarization coherence maps, magnitude imagery, and polarimetric decompositions. Residual defocus, due to platform motion during data acquisition, was corrected with a digital elevation model-based autofocus algorithm. We generated and exploited the VideoSAR image products to characterize the surface movement effects caused by the underground explosion. Unlike seismic sensors, which measure local area seismic waves using sparse spacing and subterranean positioning, these VideoSAR products captured high-spatial resolution, 2-D, time-varying surface movement. The results from the fifth SPE (SPE-5) used single-polarimetric VideoSAR data. In this paper, we present single-polarimetric and fully polarimetric VideoSAR results while monitoring the SPE-6 underground chemical explosion. We show that fully polarimetric VideoSAR imaging provides a unique, coherent, time-varying measure of the surface expression of the SPE-6 underground chemical explosion. We include new surface characterization results from the measured PolSAR SPE-6 data via H/A/α polarimetric decomposition.

Scene Classification Using Hierarchical Wasserstein CNN

Wed, 05/01/2019 - 00:00
In multiclass classification, convolutional neural network (CNN) is generally coupled with the cross-entropy (CE) loss, which only penalizes the predicted probability corresponding to a ground truth class and ignores the interclass relationship. We argue that CNN can be improved by using a better loss function. On the other hand, the Wasserstein distance (WD) is a well-known metric used to measure the distance between two distributions. Directly solving the WD problem requires a prohibitively large amount of computation time, whereas the cheaper iterative algorithms have a variety of shortcomings such as computational instability and difficulty in selecting parameters. In this paper, we address these issues by giving an analytical solution to the WD problem-for the first time, we find that for two distributions in hierarchically organized data space, WD has a closed-form solution, which we call “hierarchical WD (HWD).” We use this theory to construct novel loss functions that overcome the shortcomings of CE loss. To this end, multi-CNN information fusion that provides the basis for building category hierarchies is carried out first. Then, the semantic relationship among classes is modeled as a binary tree. Then, CNN coupled with an HWD-based loss, i.e., hierarchical Wasserstein CNN (HW-CNN), is trained to learn deep features. In this way, prior knowledge about the interclass relationship is embedded into HW-CNN, and information from several CNNs provides guidance in the process of training individual HW-CNNs. We conducted extensive experiments over two publicly available remote sensing data sets and achieved a state-of-the-art performance in scene classification tasks.

Phasor Quaternion Neural Networks for Singular Point Compensation in Polarimetric-Interferometric Synthetic Aperture Radar

Wed, 05/01/2019 - 00:00
Interferograms obtained by synthetic aperture radar often include many singular points (SPs), which makes it difficult to generate an accurate digital elevation model. This paper proposes a filtering method to compensate SPs adaptively by using polarization and phase information around the SPs. Phase value is essentially related to polarization changes in scattering as well as propagation. In order to handle the polarization and phase information simultaneously in a consistent manner, we define a new number, phasor quaternion (PQ), by combining quaternion and complex amplitude, with which we construct the theory of PQ neural networks (PQNNs). Experiments demonstrate that the proposed PQNN filter compensates SPs very effectively. Even in the situations where the conventional methods deteriorate in their performance, it realizes accurate compensation, thanks to its good generalization characteristics in integrated Poincare-sphere polarization space and the complex-amplitude space. We find that PQNN is an excellent framework to deal with the polarization and phase of electromagnetic wave adaptively and consistently.

Markov Random Fields Integrating Adaptive Interclass-Pair Penalty and Spectral Similarity for Hyperspectral Image Classification

Wed, 05/01/2019 - 00:00
This paper presents a novel Markov random field (MRF) method integrating adaptive interclass-pair penalty (aICP2) and spectral similarity information (SSI) for hyper-spectral image (HSI) classification. aICP2 structurally combines K(K - 1)/2 (K is the number of classes) classical “Potts model” with K(K - 1)/2 interaction coefficients. aICP2 tries a new way to solve the key problems, insufficient correction within homogeneous regions, and over-smoothness at class boundaries, in MRF-based HSI classification. It is assumed that different class pairs should be assigned with various degrees of penalties in MRF smoothness process, according to pairwise class separability and spatial class confusion in raw classification map. The Fisher ratio is modified to measure pairwise class separability with a training set. And, gray level co-occurrence matrix is used to measure spatial class confusion degree. Then, aICP2 is constructed by combining Fisher ratio and GCLM. aICP2 applies larger penalty on class pairs that confuse with each other seriously to provide sufficient smoothness, and vice versa. In addition, to protect class edges and details, SSI is introduced to make the penalty of related neighboring pixels small. aICP2ssi denotes the integration of aICP2 and SSI. The further improved method is both interclass-pair and interpixel adaptive in spatial term. A graph-cut-based α-β-swap method is introduced to optimize the proposed energy function. The experimental results on real HSI data indicate that the proposed method outperforms compared MRF-based and other spectral-spatial approaches in terms of classification accuracies and region uniformity.

Incoherent Range Walk Compensation for Spaceborne GNSS-R Imaging

Wed, 05/01/2019 - 00:00
Global navigation satellite system reflectometry (GNSS-R) receivers produce delay-Doppler maps (DDMs) by incoherently integrating coherent integration results. Due to system dynamics, during incoherent integration, the receiver aligns each coherent result by tracking the delay and Doppler of the specular point. This is known to cause a blurring of the spatial footprint of the Woodward ambiguity function (WAF) on the reflecting surface. In this paper, we demonstrate that the blurring of the WAF varies over the glistening zone (GZ), and even if a fixed point on the ground is tracked, blurring still occurs. We derive the expressions for the delay and Doppler change rates over the GZ and then predict the error introduced by range walk for typical GNSS-R scatterometry configurations. We find that ≈6 dB of loss is expected for a point scatterer near the edge of the GZ when a fixed point on the surface is tracked. The incoherent range walk compensation (IRWC) method is then presented for GNSS-R receivers to mitigate this loss. The IRWC method focuses the power in the DDM to the isodelay and iso-Doppler configuration occurring at the midpoint of the integration time. DDMs produced by tracking a fixed point with and without IRWC are simulated, and errors are found to be in agreement with those predicted. Spatial domain GNSS-R products will be improved with IRWC. Target detection will benefit from a larger usable swath, allowing longer tracking and detection times as a result of the increased target to clutter and noise ratio. At the same time, imaging applications will no longer suffer from a spatially variant blurring of the WAF, which limits the resolution of the estimated products. IRWC is shown to mitigate the range migration losses and improve the SNR of an imaging GNSS-R receiver by ≈6 dB near the edge of the GZ.

Monostatic and Bistatic Scattering Modeling of the Anisotropic Rough Soil

Wed, 05/01/2019 - 00:00
The electromagnetic scattering generated by agricultural tilled soils can be affected by a strong anisotropic component of the rough-surface profile. An accurate and reliable modeling of the normalized radar cross section, under both monostatic and bistatic geometries, is particularly important and desirable, especially for the correct estimation of the soil moisture content by means of satellite-based observations. In this paper, moving from the modeling so far proposed in the literature, we present and discuss a novel, more general (i.e., 2-D), spectral representation of an agricultural tilled soil, implementing a solution of the scattering based on the first-order small-slope approximation. Comparisons are given with a well-established model based on a 1-D representation of the soil correlation function, accounting for the radiation pattern of the illuminating antenna. The investigation gives new insight on the phenomenology of the bistatic scattering from the anisotropic soil, providing interesting information for the next generation of satellite missions, which foresees the possibility of launching companion satellites carrying aboard a passive receiver collecting the signal transmitted by active SAR-based platforms.

Estimating Summertime Precipitation from Himawari-8 and Global Forecast System Based on Machine Learning

Wed, 05/01/2019 - 00:00
Random forests (RFs), an advanced machine learning (ML) method, was used here to develop a robust and rapid quantitative precipitation estimates (QPEs) algorithm for the new-generation geostationary satellite of Himawari-8. In this algorithm, the global precipitation measurement (GPM) product has been employed to train QPE prediction model. The real-time multiband infrared brightness temperature from Himawari-8, combined with the spatiotemporally matched numerical weather prediction (NWP) data from the global forecast system, have been used as predictor variables for QPE. Among the variables used in RF learning model, total precipitable water and K -index from NWP data have the highest rankings, indicating the importance of atmospheric environment for QPE. To enhance the accuracy of RF models or to optimize model training, a sample-balance technique has been utilized to adjust the ratios of samples in non-precipitation/precipitation classification and quantitative precipitation regression data sets. Further sensitivity and validation analyses help determine the optimal RF classification and regression models for predicting nonprecipitation/precipitation pixel and rain rate. The selected RF classification model is found to predict precipitation area with an accuracy of 0.87. For predicted QPE product, the mean-absolute-error and root-mean-square error of RF regression model are 0.51 and 2.0 mm/h, respectively. Overall, the RF ML algorithm has a higher detection rate over homogenous ocean surface as compared with over land. Meanwhile, this RF algorithm tends to underestimate rain rate, especially in the presence of heavy rainfall. Despite this, it still produces a reasonable pattern of rainfall area and intensity, which are highly consistent with GPM observations.

A New Method for Ionospheric Tomography and Its Assessment by Ionosonde Electron Density, GPS TEC, and Single-Frequency PPP

Wed, 05/01/2019 - 00:00
A new tomographic method was developed with the main goal of mapping the ionosphere in the region of Brazil. The ionospheric background was estimated based on ionosonde and radio-occultation measurements to overcome the lack of data provided by climatological models in low-latitude regions. A new method of performing iterations of the conventional multiplicative algebraic reconstruction technique (MART) was also used in order to avoid nonilluminated cells and improve the spatial resolution. The quality assessment using independent ionosonde data during two weeks in 2013 showed a better performance of the proposed method in comparison to the international reference ionosphere, providing improvements of 31% for the F-layer peak height hm and 24% for the ionospheric peak of electron density Nm. The tomographic technique was also evaluated in the estimation of the total electron content (TEC) and in the single-frequency precise point positioning (PPP). Improvements of 59% in TEC and 31% in the single-frequency PPP were obtained in comparison to the results derived from global ionospheric maps. In addition, a better daily description of the ionosphere was obtained using the proposed method, where it was possible to detect the peak height increasing associated with the prereversal enhancement of the vertical plasma drift that occurs near sunset hours. The results reveal that the modified form of the MART tomographic technique can be considered a useful tool for technical and scientific communities involved in space weather, spatial geodesy, and telecommunications.

GPR Target Detection by Joint Sparse and Low-Rank Matrix Decomposition

Wed, 05/01/2019 - 00:00
Ground penetrating radar (GPR) uses electromagnetic waves to image, locate, and identify changes in electric and magnetic properties in the ground. The received signal comprises not only the target echoes but also strong reflections from the rough, uneven ground surface, which impair subsurface inspections and visualization of buried objects. In this paper, a background clutter mitigation and target detection method using low-rank and sparse priors is proposed for GPR data. The radar signal is decomposed into the sum of a low-rank component and a sparse component, plus noise. The low-rank component captures the ground surface reflections and background clutter, whereas the sparse component contains the target reflections. The effectiveness of the proposed method is evaluated on real radar signals collected from buried landmines and improvised explosive devices. The experimental results show that the proposed method successfully removes the background clutter and estimates the target signals.

Hyperspectral Image Denoising by Fusing the Selected Related Bands

Wed, 05/01/2019 - 00:00
Hyperspectral images (HSIs) convey more useful information than RGB or gray images, which are widely used in many remote sensing tasks. In real scenarios, HSIs are inevitably corrupted by noise because of sensors' imperfectness or atmospheric influence. Recently, many HSI denoising methods have been proposed to utilize the interband information between different spectral bands. However, these methods regard the HSI as a whole and treat the different spectral bands with the same noise level. In fact, the noise levels in different bands are different. Especially, only few certain bands are corrupted by noise, named the target noised bands. Under this circumstance, an HSI denoising method is proposed by considering the band relationship and different noise levels. The target noised bands are adaptively denoised by fusing some selected bands. Specifically, some related but quality superior bands are selected according to the target noised bands. Then, the target noised bands can be denoised by fusing the selected related bands. Experimental results show that the proposed method achieves considerable performances in comparison with several state-of-the-art hyperspectral denoising methods.

Pulse-to-Pulse Correlation Effects in High PRF Low-Resolution Mode Altimeters

Wed, 05/01/2019 - 00:00
In this paper, we revisit the pulse-to-pulse correlation properties of nadir-looking pulse-limited altimeters, with the objective of determining the effect of the partial correlation of radar echoes transmitted at much higher rate than the conventional pulse repetition frequency (PRF). This is particularly relevant for the Sentinel-6/Jason-CS mission. The pulse-to-pulse echo power autocorrelation shows much shorter decorrelation times toward the trailing edge of the waveform than those observed for range gates close to the leading edge. At high PRFs this creates a significant variability in the statistical properties of the range gates in the 20-Hz multilooked waveforms. By processing an extensive data set of CryoSat-2 Synthetic Aperture Radar mode data in a pseudo-low resolution mode fashion, we determined that despite the fact that at higher PRFs the noise in the estimation of geophysical parameters is reduced, significant sea-state-dependent biases are also introduced during the retracking process, which are particularly relevant for sea surface height and significant wave height. Those biases will need to be appropriately accounted for when integrating Sentinel-6/Jason-CS data in a climatological data record.

All-Directions Through-the-Wall Imaging Using a Small Number of Moving Omnidirectional Bi-Static FMCW Transceivers

Wed, 05/01/2019 - 00:00
Through-the-wall radar imaging is a powerful tool for mapping buildings' interiors and hidden objects behind the walls. Through-the-wall imaging systems require large linear arrays of directive antennas to form a large aperture for obtaining images with a high cross-range resolution. However, the low mobility and limited field of view of the conventional systems limit their imaging capability. The concept of all-directions through-the-wall imaging has recently been proposed to enhance the mobility, cross-range resolution, and field of view of the through-the-wall imaging systems. In this technique, the large linear array of directive antennas is replaced by a dense 2-D synthetic array formed by small moving transceivers utilizing omnidirectional antennas. The 2-D synthetic array provides 360° high cross-range resolution images. This paper focuses on the implementation of a system realizing all-directions through-the-wall imaging and measurement results. A bi-static frequency-modulated continuous wave (FMCW) radar system utilizing a simple wireless synchronization scheme and wideband omnidirectional antennas is fabricated and an image formation technique compatible with bi-static FMCW imaging system is presented. Measurement results show that the imaging system can provide 360° high-resolution image of objects and walls in a short time.

Self-Supervised Feature Learning With CRF Embedding for Hyperspectral Image Classification

Wed, 05/01/2019 - 00:00
The challenges in hyperspectral image (HSI) classification lie in the existence of noisy spectral information and lack of contextual information among pixels. Considering the three different levels in HSIs, i.e., subpixel, pixel, and superpixel, offer complementary information, we develop a novel HSI feature learning network (HSINet) to learn consistent features by self-supervision for HSI classification. HSINet contains a three-layer deep neural network and a multifeature convolutional neural network. It automatically extracts the features such as spatial, spectral, color, and boundary as well as context information. To boost the performance of self-supervised feature learning with the likelihood maximization, the conditional random field (CRF) framework is embedded into HSINet. The potential terms of unary, pairwise, and higher order in CRF are constructed by the corresponding subpixel, pixel, and superpixel. Furthermore, the feedback information derived from these terms are also fused into the different-level feature learning process, which makes the HSINet-CRF be a trainable end-to-end deep learning model with the back-propagation algorithm. Comprehensive evaluations are performed on three widely used HSI data sets and our method outperforms the state-of-the-art methods.

The Challenges of Interpreting Oil–Water Spatial and Spectral Contrasts for the Estimation of Oil Thickness: Examples From Satellite and Airborne Measurements of the Deepwater Horizon Oil Spill

Wed, 05/01/2019 - 00:00
Optical remote sensing is one of the most commonly used techniques to detect oil in the surface ocean. This is because oil has optical properties that are different from water to modulate oil-water spatial and spectral contrasts. However, understanding these contrasts is challenging because of variable results from laboratory and field experiments as well as from different observing conditions and spatial/spectral resolutions of remote sensing imagery. Here, through reviewing published oil-water spectral contrasts and analyzing remotely sensed spectra collected by several satellite and airborne sensors (MERIS, MODIS, MISR, Landsat, and AVIRIS) from the Deepwater Horizon oil spill, we provide the interpretation of the spatial/spectral contrasts of various oil slicks and discuss the challenges in such interpretations. In addition to oil thickness, several other factors also affect oil-water spatial/spectral contrasts, including sun glint strength, oil emulsification state, optical properties of oil covered water, and spatial/spectral resolutions of remote sensing imagery. In the absence of high spatial- and spectral-resolution imagery, a multistep scheme may be used to classify oil type (emulsion and non-emulsion) and to estimate relative oil thickness for each type based on the known optical properties of oil, yet such a scheme requires further research to improve and validate.

Pixel-Wise MTInSAR Estimator for Integration of Coherent Point Selection and Unwrapped Phase Vector Recovery

Wed, 05/01/2019 - 00:00
Coherent point (including persistent and distributed scatterers) selection and phase ambiguity treatment (or parameter estimation) are the key tasks involved in multitemporal InSAR (MTInSAR) algorithms, which are usually conducted separately with empirical thresholds. It is not rare to see that due to the discrepancies on threshold setting, even for the same MTInSAR technique with the same data sets, it will raise different (sometimes quite notable) results and affect the applicability of InSAR techniques. We propose here an integrated MTInSAR estimator that combines the coherent point selection and phase vector unwrapping into a single step. Essentially, the estimator aims to recover the unwrapped phase vector at coherent points. Therefore, it could serve as an alternative solution of spatial-temporal phase unwrapping problem. In the estimator, wrapped phase at all pixels in short baseline interferograms are taken as observations. Starting from the phase differences at arcs of a fully connected network of pixels, based on the residual analysis and spatial closure of phase triangularity, the estimator can detect and delete the arcs having unacceptable phase noise and phase ambiguities. By integrating the phase differences at the remained arcs, the unwrapped phase at coherent points in consecutive acquisition intervals can be obtained. Impressively, the estimator is immune to the bias raised by improper deformation model. The performance of the proposed estimator is evaluated via semisynthetic and real data tests. Considering that the phase enhancement algorithms (e.g., phase-linking and Extended Minimum Cost Flow-Small BAseline Subset) that can reconstruct high-quality wrapped phases are gaining popularity, the proposed estimator can also be implemented as a postprocessing module of these algorithms for retrieval of unwrapped phase vectors at coherent points.

Unsupervised Feature Extraction in Hyperspectral Images Based on Wasserstein Generative Adversarial Network

Wed, 05/01/2019 - 00:00
Feature extraction (FE) is a crucial research area in hyperspectral image (HSI) processing. Recently, due to the powerful ability of deep learning (DL) to extract spatial and spectral features, DL-based FE methods have shown great potentials for HSI processing. However, most of the DL-based FE methods are supervised, and the training of them suffers from the absence of labeled samples in HSIs severely. The training issue of supervised DL-based FE methods limits their application on HSI processing. To address this issue, in this paper, a novel modified generative adversarial network (GAN) is proposed to train a DL-based feature extractor without supervision. The designed GAN consists of two components, which are a generator and a discriminator. The generator can focus on the learning of real probability distributions of data sets and the discriminator can extract spatial-spectral features with superior invariance effectively. In order to learn upsampling and downsampling strategies adaptively during FE, the proposed generator and discriminator are designed based on a fully deconvolutional subnetwork and a fully convolutional subnetwork, respectively. Moreover, a novel min-max cost function is designed for training the proposed GAN in an end-to-end fashion without supervision, by utilizing the zero-sum game relationship between the generator and discriminator. Besides, the proposed modified GAN replaces the original Jensen-Shannon divergence with the Wasserstein distance, aiming to mitigate the unstability and difficulty of the training of GAN frameworks. Experimental results on three real data sets validate the effectiveness of the proposed method.

Theme by Danetsoft and Danang Probo Sayekti inspired by Maksimer