Search

Estimates of the area of land cover classes or land change are frequently calculated from land cover classification maps by counting the pixels labeled as each class in the map. This procedure is known to produce biased estimates of area for many widely used classification algorithms, including random forests. Poststratification estimation using the mapped classes as strata has been proposed to obtain unbiased estimates of the class areas. Still, the method requires additional sampling units, which may not be available or be the most efficient method depending on the application. Alternatively, consistent estimates of class areas can be obtained using class membership probabilities estimates from a random forest classification. This article demonstrates that, for a large sample and proper set of explanatory variables, the error of the predicted class membership probabilities obtained from a random forest classification converges to zero. Therefore, the expected class areas calculated from these probabilities converge to the population class areas. On average, the relative error of the expected class proportions computed by class membership probabilities from a random forests model was 40% points lower than the proportions estimated by pixel counting. Our proposed approach is also comparable to the area-adjusted method, which is currently considered the best practice by the remote sensing community. We recommend that class probability estimates area always retained and used for calculating expected class areas or area proportions based on our results. Our method reduces bias compared to statistics calculated by pixel counting and circumvents the need for poststratification area estimates under certain conditions. [ABSTRACT FROM AUTHOR]

Inverse mapping from geophysics to geology is a difficult problem due to the inherent uncertainty of geophysical data and the spatially heterogeneous patterns (structure) in geology. We improve conditional facies simulation based on GANs (GANSim), a new Earth model simulation method based on generative adversarial networks (GANs), to bridge the gap between remotely sensed geophysical information and geology, by introducing a specially designed loss function and an input architecture for probability maps representing geophysical interpretation. After training, the GANSim is then used to produce multiple geological facies models conditioned to the input geophysics-interpreted probability maps alone or together with input well observations and global features. By evaluation, the generated facies models are realistic, diversified, and consistent with all input conditions. We demonstrate that the GAN learns the implicit geological pattern knowledge from the training data set and the knowledge of conditioning to inputs from human-defined explicit functions. Given the commonality of remotely sensed geophysical information, sparse measurements, and global features, GANSim should be applicable to many problems of geosciences. [ABSTRACT FROM AUTHOR]

Receiver operating characteristic (ROC) analysis is performed by a curve, called ROC curve, plotted based on detection probability, PD, versus false alarm probability, PF, and has been widely used as an evaluation tool for signal detection. Specifically, the area under an ROC curve (AUC) is calculated and used as a detection measure. Unfortunately, finding distributions of PD and PF to generate a continuous ROC curve is practically infeasible. This article investigates approaches to generating a discrete 2D ROC curve of (PD, PF) without appealing for probability distributions. Since PD and PF are determined by the same threshold τ to specify a detector, an ROC curve of (PD, PF) can only be used to evaluate the effectiveness of a detector but not target detectability (TD) and also not background suppressibility (BS). To address this issue, a 3D ROC curve is generated as a function of (PD, PF, τ) by introducing a specific threshold parameter τ as a third independent variable. As a result, a 3D ROC curve along with its derived three 2D ROC curves of (PD, PF), (PD, τ), and (PF, τ) can further be used to design new quantitative measures to evaluate the effectiveness of a detector and its TD and BS. To demonstrate the full utility of 3D ROC analysis in target detection, extensive experiments are performed on two types of targets, prior target detection and anomaly detection, to conduct a comprehensive analysis on 3D ROC curves using new designed detection measures to evaluate target/anomaly detection performance. [ABSTRACT FROM AUTHOR]

A new technique for automated retrieval of ice concentration from RADARSAT-2 dual-polarization HH-HV ScanSAR Wide images for subsequent assimilation in ice numerical models is presented. First, we extended our previously introduced ice and water detection approach operating at a 2.05 km $\times2.05$ km spatial scale to a set of 19 different spatial scales ranging from 2.05 km (41 pixels) down to 0.25 km (5 pixels). As the spatial resolution was increased, the overall accuracy of ice and water detection stayed at a very high level across all scales (between 99.5% and 99.8%), but the number of water retrievals substantially dropped. Second, we designed an approach for estimating ice concentration in a 2 km $\times 2$ km ($40\times40$ pixels) area consisting of $64\,\,5 \times 5$ pixel blocks. The $5\times 5$ pixel blocks which are initially classified as unknowns are iteratively combined in clusters with effective spatial scales larger than 5 pixels. The clusters are further classified as ice or water using the ice probability model corresponding to the effective spatial scale. The $40\times40$ pixel area becomes populated with high-resolution ($5\times 5$ pixels) ice and water retrievals, and the ice concentration is estimated based on the number of ice and water retrievals. The proposed approach produces a much better agreement with the Canadian Ice Service Image Analysis ice concentrations (root-mean-square error (RMSE) = 2.2%) compared to the original 2-km ice/water detection approach (RMSE = 19.9%). The developed technique will be adapted to the RADARSAT Constellation Mission data for data assimilation in Environment and Climate Change Canada Regional Ice-Ocean Prediction System. [ABSTRACT FROM AUTHOR]

In this paper, we present a new technique for automated detection of ice and open water from sequential RADARSAT-2 ScanSAR dual-polarization HH–HV images. The technique is based on combining a previously developed approach to ice and water detection applied to single synthetic aperture radar (SAR) images with the ice motion information derived from sequential SAR images. To evaluate the new approach, it was applied to 736 SAR image pairs acquired in 2013. Compared with the previous approach, the new approach produced an increase in the fraction of correctly classified water samples from 57.7% to 72.6% while the fraction of correctly classified ice samples did not change appreciably. The overall accuracy stayed at a high level exceeding 99.8%, when compared against the Canadian Ice Service Image Analysis pure ice and water samples. Verification results for different regions and months showed that the detection accuracy exceeds 99.5% for the most regions and months. The proposed approach can also assign enhanced quality to ice and water retrievals found in the reference image. The results are particularly relevant in light of the upcoming Canadian RADARSAT Constellation Mission which will significantly increase the amount and frequency of SAR observations over the Arctic region. [ABSTRACT FROM AUTHOR]

This paper presents a statistical detection theory approach to hyperspectral image (HSI) classification which is quite different from many conventional approaches reported in the HSI classification literature. It translates a multi-target detection problem into a multi-class classification problem so that the well-established statistical detection theory can be readily applicable to solving classification problems. In particular, two types of classification, a priori classification and a posteriori classification, are developed in corresponding to Bayes detection and maximum a posteriori (MAP) detection, respectively, in detection theory. As a result, detection probability and false alarm probability can also be translated to classification rate and false classification rate derived from a confusion classification matrix used for classification. To evaluate the effectiveness of a posteriori classification, a new a posteriori classification measure, to be called precision rate (PR), is also introduced by MAP classification in contrast to overall accuracy (OA) that can be considered as a priori classification measure and has been used for Bayes classification. The experimental results provide evidence that a priori classifier as Bayes classifier which performs well in terms of OA does not necessarily perform well as a posteriori classifier in terms of PR. That is, PR is the only criterion that can be used as a posteriori classification measure to evaluate how well a classifier performs. [ABSTRACT FROM AUTHOR]

A new directional canopy emissivity model (CE-P) based on spectral invariants is proposed in this paper. First, we prove the existence of the spectral invariant properties in the thermal infrared (TIR) band using a Monte Carlo model. Based on it, the equation of the new model is derived from the perspective of absorption. In this expression, single-scattering and multiscattering effects are separated analytically in the TIR band. We find that the overall contribution of multiple scatterings is less than 0.005 when the component emissivities are over 0.90, and the overall contribution decreases with increasing leaf or soil emissivity. Furthermore, the new model can avoid the logical difficulty encountered when using the traditional cavity effect factor to simulate the emissivity of a sparse vegetation canopy. The results of 4SAIL and Discrete Anisotropic Radiative Transfer (DART) are selected to do cross validation. The CE-P can achieve a high accuracy compared with 4SAIL and DART, with an absolute bias less than 0.002 when the leaf (soil) emissivity is equal to 0.98 (0.94). Four widely used analytical models are selected for comparison. The resulting accuracies of these models are ordered from CE-P to REN15, FR97, FR02, and VALOR96 with the most serious error up to 0.002, 0.002, 0.007, 0.013, and 0.014, respectively. Three main conclusions are obtained through the sensitivity analysis: the multiscattering between vegetation and the background can be ignored when the leaf (soil) emissivity is no less than 0.94 (0.90), the second and higher order scattering within the vegetation can also be ignored when the leaf (soil) emissivity is no less than 0.94 (0.90), and the single-scattering effect within the canopy should be considered which can be calculated using three view factors. [ABSTRACT FROM AUTHOR]

Probabilistic flood mapping offers flood managers, decision makers, insurance agencies, and humanitarian relief organizations a useful characterization of uncertainty in flood mapping delineation. Probabilistic flood maps are also of high interest for data assimilation into numerical models. The direct assimilation of probabilistic flood maps into hydrodynamic models would be beneficial because it would eliminate the intermediate step of having to extract water levels first. This paper introduces a probabilistic flood mapping procedure based on synthetic aperture radar (SAR) data. Given a SAR image of backscatter values, we construct a total histogram of backscatter values and decompose this histogram into probability distribution functions of backscatter values associated with flooded (open water) and non-flooded pixels, respectively. These distributions are then used to estimate, for each pixel, its probability of being flooded. The new approach improves on binary SAR-based flood mapping procedures, which do not inform on the uncertainty in the pixel state. The proposed approach is tested using four SAR images from two floodplains, i.e., the Severn River (U.K.) and the Red River (U.S.). In all four test cases, reliability diagrams, with error values ranging from 0.04 to 0.23, indicate a good agreement between the SAR-derived probabilistic flood map and an independently available validation map, which is obtained from aerial photography. [ABSTRACT FROM PUBLISHER]

Classification of hyperspectral images usually suffers from high dimensionality and few reference data, which limits the performance of the pixelwise classifiers. The spectral–spatial classifiers, which integrate the spectral data and the spatial information during the classification, perform impressively in terms of the high classification accuracy and the homogeneous appearance of the classification map. In this paper, we propose a new probabilistic framework for spectral–spatial classification (PFSSC), which integrates the spectral data and the spatial information from the probabilistic point of view. Both the spectral data and the spatial information are used to estimate the per-pixel probability, which gives the likelihood that one pixel belongs to one class, respectively. The classification map can then be directly derived from the joint probability. In the proposed framework, a pixelwise probabilistic classifier can be extended as a spectral–spatial one since it can integrate spatial information easily. Furthermore, these spectral–spatial classifiers in the proposed framework are realized in an iterative way to avoid the problem caused by the limited reference data to some extent. In each iterative step, some unassigned pixels are classified by considering the pixels assigned in previous iterative steps. In this iterative process, pixels are assigned to specific labels step by step gradually. In the proposed framework, the probabilistic support vector machine (SVM) and random forest (RF) are extended to be two spectral–spatial classifiers. In short, we denote them as SVM-PFSSC and RF-PFSSC, respectively. The experimental results show that SVM-PFSSC and RF-PFSSC outperform some pixelwise and spectral–spatial classifiers. [ABSTRACT FROM AUTHOR]

Scene classification has been proved to be an effective method for high spatial resolution (HSR) remote sensing image semantic interpretation. The probabilistic topic model (PTM) has been successfully applied to natural scenes by utilizing a single feature (e.g., the spectral feature); however, it is inadequate for HSR images due to the complex structure of the land-cover classes. Although several studies have investigated techniques that combine multiple features, the different features are usually quantized after simple concatenation (CAT-PTM). Unfortunately, due to the inadequate fusion capacity of \boldsymbolk-means clustering, the words of the visual dictionary obtained by CAT-PTM are highly correlated. In this paper, a semantic allocation level (SAL) multifeature fusion strategy based on PTM, namely, SAL-PTM (SAL-pLSA and SAL-LDA) for HSR imagery is proposed. In SAL-PTM: 1) the complementary spectral, texture, and scale-invariant-feature-transform features are effectively combined; 2) the three features are extracted and quantized separately by \boldsymbolk-means clustering, which can provide appropriate low-level feature descriptions for the semantic representations; and 3)the latent semantic allocations of the three features are captured separately by PTM, which follows the core idea of PTM-based scene classification. The probabilistic latent semantic analysis (pLSA) and latent Dirichlet allocation (LDA) models were compared to test the effect of different PTMs for HSR imagery. A U.S. Geological Survey data set and the UC Merced data set were utilized to evaluate SAL-PTM in comparison with the conventional methods. The experimental results confirmed that SAL-PTM is superior to the single-feature methods and CAT-PTM in the scene classification of HSR imagery. [ABSTRACT FROM PUBLISHER]

We present a semisupervised method for single-pixel classification of hyperspectral images. The proposed method is designed to address the special problematic characteristics of hyperspectral images, namely, high dimensionality of hyperspectral pixels, lack of labeled samples, and spatial variability of spectral signatures. To alleviate these problems, the proposed method features the following components. First, being a semisupervised approach, it exploits the wealth of unlabeled samples in the image by evaluating the confidence probability of the predicted labels, for each unlabeled sample. Second, we propose to jointly optimize the classifier parameters and the dictionary atoms by a task-driven formulation, to ensure that the learned features (sparse codes) are optimal for the trained classifier. Finally, it incorporates spatial information through adding a Laplacian smoothness regularization to the output of the classifier, rather than the sparse codes, making the spatial constraint more flexible. The proposed method is compared with a few comparable methods for classification of several popular data sets, and it produces significantly better classification results. [ABSTRACT FROM PUBLISHER]

In this paper, we present a method for the multitemporal and contextual classification of georeferenced optical remote sensing images acquired at different epochs and having different geometrical resolutions. The method is based on Conditional Random Fields (CRFs) for contextual classification. The CRF model is expanded by temporal interaction terms that link neighboring epochs via transition probabilities between different classes. In order to be able to deal with data of different resolution, the class structure at different epochs may vary with the resolution. The goal of the multitemporal classification is an improved classification performance at all individual epochs, but also the detection of land-cover changes, possibly using lower resolution data. This paper also contains a comparison of the performance of different models for the interaction potentials. Results are given for two different test sites in Germany, where Ikonos, RapidEye, and Landsat images are available. Our results show that the multitemporal classification does indeed increase the overall accuracy of all epochs compared to a monotemporal classification and to a state-of-the-art multitemporal classification method, and that it is feasible to detect changes in lower resolution images. [ABSTRACT FROM PUBLISHER]

Delay/Doppler altimetry (DDA) aims at reducing the measurement noise and increasing the along-track resolution in comparison with conventional pulse-limited altimetry. In a previous paper, we have proposed a semi-analytical model for DDA, which considers some simplifications as the absence of mispointing antenna. This paper first proposes a new analytical expression for the flat surface impulse response (FSIR), considering antenna mispointing angles, a circular antenna pattern, no vertical speed effect, and uniform scattering. The 2-D delay/Doppler map is then obtained by a numerical computation of the convolution between the proposed analytical function, the probability density function of the heights of the specular scatterers, and the time/frequency point target response of the radar. The approximations used to obtain the semi-analytical model are analyzed, and the associated errors are quantified by analytical bounds for these errors. The second contribution of this paper concerns the estimation of the parameters associated with the multilook semi-analytical model. Two estimation strategies based on the least squares procedure are proposed. The proposed model and algorithms are validated on both synthetic and real waveforms. The obtained results are very promising and show the accuracy of this generalized model with respect to the previous model assuming zero antenna mispointing. [ABSTRACT FROM PUBLISHER]

Synthetic aperture radar (SAR) images of ship wakes frequently exhibit long streaks. Some of the streaks are attributed to the Kelvin V-wave wake, and others are attributed to the turbulent and bright narrow V-wake. Ocean waves reduce the detectability of Kelvin wakes in SAR images. The SAR imaging of Kelvin wakes of two ships with different dimensions moving in rough sea is studied theoretically. The Kelvin wakes are simulated using the Michell theory. The random sea surface is simulated using the Pierson-Moskowitz type wave spectrum. The elevations of the points of the water surface and their velocities and accelerations are considered as a superposition of a ship's coherent waves and wind waves. A two-scale composite model and the velocity-bunching integral are used to simulate the intensity of SAR images of a disturbed sea surface with an embedded Kelvin wake. The fast discrete Radon transform is applied to detect the boundaries of the simulated Kelvin wakes. The probabilities of missing detections (PMD) and false alarms (PFA) are estimated numerically. The influence of ships, wind, and SAR parameters on PMD and PFA is discussed. It is illustrated that PMD/PFA strongly depend on the sea state. [ABSTRACT FROM PUBLISHER]

An approach was developed to retrieve the ocean slope bispectrum, which describes the nonlinearity of the slope surface, from ocean sunglint data. The departure from Gaussianity of the ocean slope was described using an N-dimensional slope joint probability density function, which was derived using a perturbative approach. The resulting Edge-worth series had various slope cumulants and cumulant functions as series coefficients, and multidimensional Hermite polynomials as the series basis functions. The slope probability density was used to specify a series of relationships between the slope and glint cumulants and cumulant functions up to third order. These relationships were inverted to retrieve the slope third cumulant function, from which we obtained the slope bispectrum via Fourier transformation. The retrieval method was validated using synthetic 1-D ocean wave slope datasets with controlled phase correlations imposed on a subset of the wave spectrum components. [ABSTRACT FROM PUBLISHER]

In this study, a novel methodology based upon the information-theoretic measures of entropy and mutual information was implemented to downscale soil moisture (SM) observations from 10 km to 1 km. It included a transformation function that related auxiliary remotely sensed (RS) products at high resolution to in situ SM observations to obtain first estimates of SM at 1 km and merging this estimate with SM at coarse resolutions through Principle of Relevant Information (PRI). The PRI-based estimates were evaluated using synthetic observations in NC Florida for heterogeneous agricultural land covers (LC), with two growing seasons of sweet corn and one of cotton, annually. The cumulative density function showed an overall error in SM of <; 0.03 cubic meter/cubic meter in the region, with a confidence interval of 95% during the simulation period. The PRI estimates at 1 km were also compared with those from the method based upon Universal Triangle (UT). The spatially averaged root mean square error (RMSE) aggregated over the vegetative LC were 0.01 cubic meter/cubic meter and 0.15 cubic meter/cubic meter using the PRI and UT methods, respectively. The RMSE for downscaled estimates using the UT method increased to 0.28 cubic meter/cubic meter when Laplacian errors are used, while the corresponding RMSE for the PRI remains the same for both Laplacian or Gaussian errors. The Kullback-Liebler divergence (KLD) for estimates using PRI is about 50% lower than those using the method based upon UT indicating that the probability density function (PDF) of the PRI estimate is closer to PDF of the true SM, than the UT method. [ABSTRACT FROM PUBLISHER]

Indicator cokriging (ICK) has been shown to be an effective subpixel mapping (SPM) algorithm. It is noniterative and involves few parameters. The original ICK-based SPM method, however, requires the semivariogram of land cover classes from prior information, usually in the form of fine spatial resolution training images. In reality, training images are not always available, or laborious work is needed to acquire them. This paper aims to seek spatial structure information for ICK when such prior land cover information is not obtainable. Specifically, the fine spatial resolution semivariogram of each class is estimated by the deconvolution process, taking the coarse spatial resolution semivariogram extracted from the class proportion image as input. The obtained fine spatial resolution semivariogram is then used to estimate class occurrence probability at each subpixel with the ICK method. Experiments demonstrated the feasibility of the proposed ICK with the deconvolution approach. It obtains comparable SPM accuracy to ICK that requires semivariogram estimated from fine spatial resolution training images. The proposed method extends ICK to cases where the prior spatial structure information is unavailable. [ABSTRACT FROM PUBLISHER]

This paper introduces a novel spectral-spatial classification method for hyperspectral images based on extended random walkers (ERWs), which consists of two main steps. First, a widely used pixelwise classifier, i.e., the support vector machine (SVM), is adopted to obtain classification probability maps for a hyperspectral image, which reflect the probabilities that each hyperspectral pixel belongs to different classes. Then, the obtained pixelwise probability maps are optimized with the ERW algorithm that encodes the spatial information of the hyperspectral image in a weighted graph. Specifically, the class of a test pixel is determined based on three factors, i.e., the pixelwise statistics information learned by a SVM classifier, the spatial correlation among adjacent pixels modeled by the weights of graph edges, and the connectedness between the training and test samples modeled by random walkers. Since the three factors are all well considered in the ERW-based global optimization framework, the proposed method shows very good classification performances for three widely used real hyperspectral data sets even when the number of training samples is relatively small. [ABSTRACT FROM PUBLISHER]

The recently developed theory of compressive sensing (CS) exhibits enormous potentials in signal recovery. In this paper, we investigate its application on spectral unmixing, which appears in hyperspectral data analysis and is usually based on a linear mixture model (LMM) that assumes that a mixed pixel is a linear combination of a set of pure spectral signatures (called endmembers) weighted by their corresponding abundances. Unlike the classical LMM that is a compact representation, we first extend it to a sparse representation (SR) by using a redundant and known endmember set instead of the complete one. Then, the SR model is multiplied by a random Gaussian measurement matrix, so spectral unmixing is casted in the framework of CS. Finally, the ℓ1-minimization algorithms are used to recover the nonnegative abundances by solving the SR model and our proposed model named CS+SR, respectively. Experimental results on both simulated and real hyperspectral data demonstrate that the CS+SR model, formed by multiplying a random Gaussian matrix on the SR model, can improve, at least in the sense of probability, the ability of the ℓ1-minimization algorithms for recovering the nonnegative sparse abundances. [ABSTRACT FROM PUBLISHER]

For many multidimensional applications of synthetic aperture radar (SAR) imaging, the estimation of the covariance matrix for each resolution cell is a critical processing step. The context of this work is the application of covariance matrix estimation for multi-baseline interferometric SAR data sets. In order to ensure local stationarity, which is needed for an unbiased estimation, adaptive techniques are necessary. In this paper, a new approach for adaptive covariance matrix estimation is proposed and evaluated based on measures known from the field of image processing. The procedure is centered around the idea of checking whether the neighboring pixels belong to the same statistical distribution as the currently investigated pixel by applying a threshold to the respective probability density function. All inlier pixels are then used to estimate the complex covariance matrix of the reference pixel. From this covariance matrix, both amplitude and interferometric phase values are extracted, which are then combined for all pixels in the stack in order to employ techniques for the evaluation of filtering efficiency that are typically used in image denoising research. It is found that the proposed algorithm provides high filtering efficiency and good detail preservation at the same time. Apart from that, it is found to be particularly suitable for small-sized stacks of coregistered SAR imagery. [ABSTRACT FROM PUBLISHER]

This paper develops several new strategies for remote sensing image classification postprocessing (CPP) and conducts a systematic study in this area. CPP is defined as a refinement of the labeling in a classified image in order to enhance its original classification accuracy. The current mainstream classification methods (preprocessing) extract additional spatial features in order to complement spectral information and enhance classification using spectral responses alone. On the other hand, however, the CPP methods, providing a new solution to improve classification accuracy by refining the initial result, have not received sufficient attention. They have potential for achieving comparable accuracy to the preprocessing methods but in a more direct and succinct way. In this paper, we consider four groups of CPP strategies: filtering; random field; object-based voting; and relearning. In addition to the state-of-the-art CPP algorithms, we also propose a series of new ones, e.g., anisotropic probability diffusion and primitive cooccurrence matrix. In experiments, a number of multisource remote sensing data sets are used for evaluation of the considered CPP algorithms. It is shown that all the CPP strategies are capable of providing more accurate results than the raw classification. Among them, the relearning approaches achieve the best results. In addition, our relearning algorithms are compared with the state-of-the-art spectral-spatial classification. The results obtained further verify the effectiveness of CPP in different remote sensing applications. [ABSTRACT FROM PUBLISHER]

Remotely sensed hyperspectral image classification is a very challenging task. This is due to many different aspects, such as the presence of mixed pixels in the data or the limited information available a priori. This has fostered the need to develop techniques able to exploit the rich spatial and spectral information present in the scenes while, at the same time, dealing with mixed pixels and limited training samples. In this paper, we present a new spectral-spatial classifier for hyperspectral data that specifically addresses the issue of mixed pixel characterization. In our presented approach, the spectral information is characterized both locally and globally, which represents an innovation with regard to previous approaches for probabilistic classification of hyperspectral data. Specifically, we use a subspace-based multinomial logistic regression method for learning the posterior probabilities and a pixel-based probabilistic support vector machine classifier as an indicator to locally determine the number of mixed components that participate in each pixel. The information provided by local and global probabilities is then fused and interpreted in order to characterize mixed pixels. Finally, spatial information is characterized by including a Markov random field (MRF) regularizer. Our experimental results, conducted using both synthetic and real hyperspectral images, indicate that the proposed classifier leads to state-of-the-art performance when compared with other approaches, particularly in scenarios in which very limited training samples are available. [ABSTRACT FROM PUBLISHER]

For inverse synthetic aperture radar imagery, the inherent sparsity of the scatterers in the range-Doppler domain has been exploited to achieve a high-resolution range profile or Doppler spectrum. Prior to applying the sparse recovery technique, preprocessing procedures are performed for the minimization of the translational-motion-induced Doppler effects. Due to the imperfection of coarse motion compensation, the autofocus technique is further required to eliminate the residual phase errors. This paper considers the phase error correction problem in the context of the sparse signal recovery technique. In order to encode sparsity, a multitask Bayesian model is utilized to probabilistically formulate this problem in a hierarchical manner. In this novel method, a focused high-resolution radar image is obtained by estimating the sparse scattering coefficients and phase errors in individual and global stages, respectively, to statistically make use of the sparsity. The superiority of this algorithm is that the uncertainty information of the estimation can be properly incorporated to obtain enhanced estimation accuracy. Moreover, the proposed algorithm achieves guaranteed convergence and avoids a tedious parameter-tuning procedure. Experimental results based on synthetic and practical data have demonstrated that our method has a desirable denoising capability and can produce a relatively well-focused image of the target, particularly in low signal-to-noise ratio and high undersampling ratio scenarios, compared with other recently reported methods. [ABSTRACT FROM PUBLISHER]

A probabilistic neural-network-based feature-matching algorithm for a stereo image pair is presented in this paper, which will be useful as a constraint initializing method for further dense matching technique. In this approach, scale-invariant feature transform (SIFT) features are used to detect interest points in a stereo image pair. The descriptor which is associated with each keypoint is based on the histogram of the gradient magnitude and direction of gradients. These descriptors are the preliminary input for the matching algorithm. Using disparity range computed by visual inspection, the search area can be restricted for a given stereo image pair. Reduced search area improves the computation speed. Initial probabilities of matches are assigned to the keypoints which are considered as probable matches from the selected search area by Bayesian reasoning. The probabilities of all such matches are improved iteratively using relaxation labeling technique. Neighboring probable matches are exploited to improve the probability of best match using consistency measures. Confidence measures considering the neighborhood, unicity, and symmetry are some validation techniques which are built into the technique presented here for finding accurate matches. The algorithm is found to be effective in matching SIFT features detected in a stereo image pair with greater accuracy, and these accurate correspondences can be used in finding the fundamental matrix which encodes the epipolar geometry between the given stereo image pair. This fundamental matrix can then be used as a constraint for finding inliers that are used in matching methods for deriving dense disparity map. [ABSTRACT FROM PUBLISHER]

Superresolution mapping (SRM) is a technique for translating original coarse-resolution fractions into a fine-resolution land cover map by dividing a coarse-resolution pixel into a few finer resolution pixels and determining their class labels. SRM can be solved by considering it a maximum a posteriori principle-based classification problem and by assigning each fine pixel as the class with the highest probability. Fine-pixel class membership probabilities (CMPs) can be calculated at two different scales: at the fine scale, in which the target fine pixel is compared with other fine pixels, and at the coarse scale, in which coarse fractions are downscaled into fine-pixel probabilities. The fine-scale CMP is suitable for representing local land cover features but not for maintaining global features. The coarse-scale CMP is the opposite of the fine-scale CMP. This paper proposes a novel multiscale approach to overcome this shortcoming by fusing the CMP calculated at both fine and coarse scales with the tau model. With the fused CMP, a simulated-annealing algorithm is applied to produce a fine-resolution land cover map. The land cover maps generated from QuickBird and IKONOS images and the National Land Cover Database were used to validate the effectiveness of the proposed SRM algorithms. The proposed SRM algorithms were evaluated visually and quantitatively by comparing them with several existing SRM algorithms. The results indicate that the accuracy of land cover maps at fine spatial resolution increased significantly compared with that obtained from all existing SRM algorithms. [ABSTRACT FROM PUBLISHER]

The concept of delay/Doppler (DD) altimetry (DDA) has been under study since the mid-1990s, aiming at reducing the measurement noise and increasing the along-track resolution in comparison with the conventional pulse-limited altimetry. This paper introduces a new model for the mean backscattered power waveform acquired by a radar altimeter operating in synthetic aperture radar mode, as well as an associated least squares (LS) estimation algorithm. As in conventional altimetry (CA), the mean power can be expressed as the convolution of three terms: the flat surface impulse response (FSIR), the probability density function of the heights of the specular scatterers, and the time/frequency point target response of the radar. An important contribution of this paper is to derive an analytical formula for the FSIR associated with DDA. This analytical formula is obtained for a circular antenna pattern, no mispointing, no vertical speed effect, and a uniform scattering. The double convolution defining the mean echo power can then be computed numerically, resulting in a 2-D semi-analytical model called the DD map (DDM). This DDM depends on three altimetric parameters: the epoch, the sea surface wave height, and the amplitude. A multi-look model is obtained by summing all the reflected echoes from the same along-track surface location of interest after applying appropriate delay compensation (range migration) to align the DDM on the same reference. The second contribution of this paper concerns the estimation of the parameters associated with the multi-look semi-analytical model. An LS approach is investigated by means of the Levenberg-Marquardt algorithm. Simulations conducted on simulated altimetric waveforms allow the performance of the proposed estimation algorithm to be appreciated. The analysis of Cryosat-2 waveforms shows an improvement in parameter estimation when compared to the CA. [ABSTRACT FROM PUBLISHER]

An indicator kriging (IK) approach for remote sensing image classification is proposed. By introducing indicator variables for categorical data, the work of image classification is transformed into estimation of class-dependent probabilities in feature space using ordinary kriging. Individual pixels are then assigned to the class with maximum class probability. The approach is distribution free and yields perfect classification accuracies for training data provided that collocated data in feature space do not exist. Technical considerations regarding implementation of IK such as indicator semivariogram modeling and handling of collocated data in feature space are also described. The IK, Gaussian-based maximum likelihood, nearest neighbor, and support vector machine (SVM) classifiers were applied to study areas within the Shimen reservoir watershed (case A: FORMOSAT-2) and Taipei city (case B: SPOT 4). The results show that the overall accuracies of the proposed IK classifier and SVM can achieve higher than 97% for training data and 81% for testing data. (The overall accuracies of IK are a little higher than those of SVM.) IK and SVM are found to be superior to the other two classifiers in terms of overall accuracies for both training and testing data. The proposed IK classifier has the following advantages: 1) It can deal with anisotropic problem in feature space; 2) it is a nonparametric method and needs not to know the type of probability distribution; and 3) it yields 100% classification accuracy for the training data provided that collocated data in feature space do not exist. [ABSTRACT FROM PUBLISHER]

The mean squared error of the classical maximum likelihood time-of-flight (ToF) estimator increases dramatically when the signal-to-noise ratio falls below a certain threshold. For narrow-band signals, this well-known threshold effect occurs largely due to the biased outliers which are induced by the local maxima of the source signal autocorrelation function. In our previous work (Part I), we have described a machine learning biosonar-inspired method for reducing the effect of the outlier bias on the accuracy of a single-echo estimate. In this paper, we extend this approach by introducing a method for combining multiple echo signals into a robust ToF estimator which is resilient to outliers. The individual bias-corrected estimates are combined together using the optimal weighted averaging (OWA) scheme which takes into account the uncertainties associated with inlier and outlier measurements. As in the single-echo case, a weak classifier and a bank of phase-shifted unmatched filters are used for estimating the probabilities of appearance of a specific outlier class in a single-echo estimate. Combined with the previously introduced single-echo bias-reduction method, the OWA scheme results in significant improvement in the ToF estimation accuracy. [ABSTRACT FROM PUBLISHER]

Accurate and timely identification of crop types has significant economic, agricultural, policy, and environmental applications. The existing remote sensing methods to identify crop types rely on remotely sensed images of high temporal frequency in order to utilize phenological changes in crop reflectance characteristics. However, these image sets generally have relatively low spatial resolution. This tradeoff makes it difficult to classify remotely sensed images in fragmented landscapes where field sizes are smaller than the resolution of imaging sensor. Here, we develop a method for combining high spatial resolution (high-resolution) data with images with low spatial resolution but with high time frequency to achieve a superior classification of crop types. The solution is implemented and tested on both synthetic and real data sets as a proof of concept. We show that, by incorporating high-temporal-frequency but low spatial resolution data into the classification process, up to 20% of improvement in classification accuracy can be achieved even if very few high-resolution images are available for a location. This boost in accuracy is roughly equivalent to including an additional high-resolution image to the temporal stack during the classification process. The limitations of the current algorithm include computational performance and the need for ideal crop curves. Nevertheless, the resulting boost in accuracy can help researchers create superior crop type classification maps, thereby creating the opportunity to make more informed decisions. [ABSTRACT FROM PUBLISHER]

Classification of remote sensing multispectral data is important for segmenting images and thematic mapping and is generally the first step in feature extraction. Per-pixel classification, based on spectral information alone, generally produces noisy classification results. The introduction of spatial information has been shown to be beneficial in removing most of this noise. Probabilistic label relaxation (PLR) has proved to be advantageous using second-order statistics; here, we present a modified contextual probabilistic relaxation method based on imposing directional information in the joint probability with third-order statistics. The proposed method was tested in synthetic images and real images; the results are compared with a “Majority” algorithm and the classical PLR method. The proposed third-order method gives the best results, both visually and numerically. [ABSTRACT FROM AUTHOR]

Road network detection from very high resolution satellite and aerial images has diverse and important usage areas such as map generation and updating. Although an expert can label road pixels in a given image, this operation is prone to errors and quite time consuming. Therefore, an automated system is needed to detect the road network in a given satellite or aerial image in a robust manner. In this paper, we propose such a novel system. Our system has three main modules: probabilistic road center detection, road shape extraction, and graph-theory-based road network formation. These modules may be used sequentially or interchangeably depending on the application at hand. To show the strengths and weaknesses of our system, we tested it on several very high resolution satellite (Geoeye, Ikonos, and QuickBird) and aerial image sets. We compared our system with the ones existing in the literature. We also tested the sensitivity of our system to different parameter values. Obtained results indicate that our system can be used in detecting the road network on such images in a reliable and fast manner. [ABSTRACT FROM AUTHOR]

This paper describes use of Advanced Scatterometer (ASCAT) C-band scatterometer data for determination of the sea ice edge. The variation in backscatter with measurement geometry is different for the sea ice surface compared to the open water surface. Utilizing the ASCAT antenna configuration with three different look angles for the same surface spot, a new ASCAT sea ice parameter has been defined. One year of ASCAT measurements has been collocated with background sea ice information to derive probability distributions for the ASCAT sea ice parameter given the known ice condition. The result can be used in an inverse methodology, a Bayesian approach, to calculate the probability of sea ice from the ASCAT measurements. The method has been tested for a full year and validated against high-resolution satellite images and ice charts from an operational Ice Service. It reveals a realistic ice edge result, however, with weather-induced noise problems in terms of “false sea ice.” For automatic ice edge detection, ancillary information is needed to remove this noise. The paper shows how the method also can be used for ice edge detection with passive microwave data from SSM/I, and how it can be extended to utilize ASCAT together with SSM/I in a multisensor approach. [ABSTRACT FROM AUTHOR]

The multiple instance regression (MIR) problem arises when a data set is a collection of bags, where each bag contains multiple instances sharing the identical real-valued label. The goal is to train a regression model that can accurately predict label of an unlabeled bag. Many remote sensing applications can be studied within this setting. We propose a novel probabilistic framework for MIR that represents bag labels with a mixture model. It is based on an assumption that each bag contains a prime instance which is responsible for the bag label. An expectation-maximization algorithm is proposed to maximize the likelihood of the mixture model. The mixture model MIR framework is quite flexible, and several existing MIR algorithms can be described as its special cases. The proposed algorithms were evaluated on synthetic data and remote sensing data for aerosol retrieval and crop yield prediction. The results show that the proposed MIR algorithms achieve higher accuracy than the previous state of the art. [ABSTRACT FROM AUTHOR]

Theories of evidence have already been applied more or less successfully in the fusion of remote sensing images. These attempts were based on the classical evidential reasoning which works under the condition that all sources of evidence and their fusion results are related to the same invariable (static) frame of discernment. When working with multitemporal remote sensing images, some change occurrences are possible between two images obtained at a different period of time, and these changes need to be detected efficiently in particular applications. The classical evidential reasoning is adapted for working with an invariable frame of discernment over time, but it cannot efficiently detect nor represent the occurrence of change from heterogeneous remote sensing images when the frame is possibly changing over time. To overcome this limitation, dynamic evidential reasoning (DER) is proposed for the sequential fusion of multitemporal images. A new state-transition frame is defined in DER, and the change occurrences can be precisely represented by introducing a statetransition operator. Two kinds of dynamical combination rules working in the free model and in the constrained model are proposed in this new framework for dealing with the different cases. Moreover, the prior probability of state transitions is taken into account, and the link between DER and Dezert-Smarandache theory is presented. The belief functions used in DER are defined similarly to those defined in the Dempster-Shafer theory. As shown in the last part of this paper, DER is able to estimate efficiently the correct change detections as a postprocessing technique. Two applications are given to illustrate the interest of DER: The first example is based on a set of two SPOT images acquired before and after a flood, and the second example uses three QuickBird images acquired during an earthquake event. [ABSTRACT FROM AUTHOR]

Magnetic anomaly detection (MAD) is a passive method used to detect visually obscured ferromagnetic objects by revealing the anomalies in the ambient Earth magnetic field. In this paper, we propose a method for MAD employing the high-order crossing (HOC) approach, which relies on the magnetic background nature. HOC is an alternative method for spectral analysis using zero-crossing count, also enabling signal discrimination. Tests with real-world recorded magnetic signals show high detection probability even for low signal-to-noise ratio. The high detection probability, together with a simple implementation and low power consumption, makes the HOC method attractive for real-time MAD applications such as intruder detection and for research on an earthquake magnetic precursor. [ABSTRACT FROM AUTHOR]

The precision of georeferencing of pushbroom satellite images without using ground control points is estimated. The study includes what-if scenarios for various precisions of star trackers, GPS receivers, image acquisition timing, interior camera parameters, and digital elevation model (DEM). Moreover, the effects of satellite altitude and sensor positions on the pushbroom array are also taken into account. The probability distribution function of the input parameters is determined by considering the sensor characteristics, environmental effects, and assumptions. The precision of the georeferenced coordinates is estimated by Monte Carlo simulation and differential sensitivity analysis. The analyses provide the most probable georeferenced coordinates and their precision for a certain precision of sensors and imaging geometry. The study is conducted based on the BilSAT orbit and SPOT 5 and Landsat optical system characteristics. An Shuttle Radar Topography Mission (SRTM) DEM and a DEM obtained from stereo images are used in georeferencing. [ABSTRACT FROM AUTHOR]

This paper proposes a new Markov-chain-based constant false alarm rate (CFAR) detector for polarimetric data using low-level data fusion and high-level decision fusion. The Markov-chain-based CFAR detector extends traditional probability density function (pdf) based CFAR detection to first-order Markov chain model by considering both correlation between neighboring pixels and pdf information in CFAR detection. With the additional correlation information, the proposed approach results in advancing the performance of conventional CFAR detectors. Moreover, to take advantage of full polarizations of polarimetric data, various data fusion methods are considered to improve detection performance, including polarimetric transformation, principal component analysis, and decision fusion. Our experimental results confirm the superiority of the new Markov chain polarimetric CFAR detector over conventional pdf-based CFAR detectors. [ABSTRACT FROM AUTHOR]

Given a set of high-resolution images of a scene, it is often desirable to predict the scene's appearance from viewpoints not present in the original data for purposes of change detection. When significant 3-D relief is present, a model of the scene geometry is necessary for accurate prediction to determine surface visibility relationships. In the absence of an a priori high-resolution model (such as those provided by LIDAR), scene geometry can be estimated from the imagery itself. These estimates, however, cannot, in general, be exact due to uncertainties and ambiguities present in image data. For this reason, probabilistic scene models and reconstruction algorithms are ideal due to their inherent ability to predict scene appearance while taking into account such uncertainties and ambiguities. Unfortunately, existing data structures used for probabilistic reconstruction do not scale well to large and complex scenes, primarily due to their dependence on large 3-D voxel arrays. The work presented in this paper generalizes previous probabilistic 3-D models in such a way that multiple orders of magnitude savings in storage are possible, making high-resolution change detection of large-scale scenes from high-resolution aerial and satellite imagery possible. Specifically, the inherent dependence on a discrete array of uniformly sized voxels is removed through the derivation of a probabilistic model which represents uncertain geometry as a density field, allowing implementations to efficiently sample the volume in a nonuniform fashion. [ABSTRACT FROM AUTHOR]

We address the problem of resolving two closely spaced complex-valued points from N irregular Fourier do- main samples. Although this is a generic super-resolution (SR) problem, our target application is SAR tomography (TomoSAR), where typically the number of acquisitions is N = 10 - 100 and SNR = 0-10 dB. As the TomoSAR algorithm, we introduce "Scale-down by LI norm Minimization, Model selection, and Estimation Reconstruction" (SL1MMER), which is a spectral estimation algorithm based on compressive sensing, model order selection, and final maximum likelihood parameter estimation. We investigate the limits of SLIMMER concerning the following questions. How accurately can the positions of two closely spaced scatterers be estimated? What is the closest distance of two scat- terers such that they can be separated with a detection rate of 50% by assuming a uniformly distributed phase difference? How many acquisitions N are required for a robust estimation (i.e., for separating two scatterers spaced by one Rayleigh resolution unit with a probability of 90%)? For all of these questions, we provide numerical results, simulations, and analytical approxima- tions. Although we take TomoSAR as the preferred application, the SLIMMER algorithm and our results on SR are generally applicable to sparse spectral estimation, including SR SAR focus- ing of point-like objects. Our results are approximately applicable to nonlinear least-squares estimation, and hence, although it is derived experimentally, they can be considered as a fundamental bound for SR of spectral estimators. We show that SR factors are in the range of 1.5-25 for the aforementioned parameter ranges of N and SNR. [ABSTRACT FROM AUTHOR]

Anomaly endmembers play an important role in the application of remote sensing, such as in unmixing classification and target detection. Inspired by the iterative error analysis (IEA), a hybrid endmember extraction algorithm (HEEA) based on a local window is proposed in this paper, which focuses on improving the accuracy of endmember extraction. HEEA uses the spectral-information-divergence–spectral-angle-distance metric to measure the similarity and the orthogonal subspace projection (OSP) method to search for the endmembers, which can decrease the correlation between extracted endmember spectra. Moreover, it is based on a local window which integrates both spatial and spectral aspects to extract endmembers. A synthetic image and Airborne Visible/Infrared Imaging Spectrometer data were tested with the HEEA method, classical IEA, OSP, simplex growing algorithm, sequential maximum angle convex cone, and spectral spatial endmember extraction automatic endmember extraction method. Experimental results indicated that HEEA manifested a slightly better improvement in the rmse and spectrum information than the other methods. The effect was investigated with various SNRs and different window sizes. The robustness of HEEA is better than the classical IEA, even with lower SNR. [ABSTRACT FROM AUTHOR]

This paper aims at defining the expression of the probability of error of classification methods using a synthetic aperture radar (SAR) intensity ratio as a classification feature. The two SAR intensities involved in this ratio can be measurements from different dates, polarizations, or, also possibly, frequency bands. Previous works provided a baseline expression of the probability of error addressing the two-class problem with equal a priori class probabilities and no calibration error. This study brings up a novel expression of the error, providing the possibility to assess the effect of class probabilities and calibration errors. An extended expression is described for the n-class problem. The effect of calibration errors such as channel gain imbalance, radiometric stability, and crosstalk is assessed in the general case. The results indicate that, for the applications under study, channel gain imbalance is usually not a decisive parameter, but radiometric stability is more critical in methods based on the temporal change. Crosstalk has a negligible effect in the case of copolarizations. The impacts of other system parameters, such as ambiguity ratio, time-lapse between repeat-pass orbits, spatial resolution, and number of looks, are illustrated through a set of assumptions on the backscattering values of the considered classes. The model is validated by comparing some of its outputs to experimental results calculated from the application of rice fields mapping methods on real data. This error model constitutes a tool for the design of future SAR missions and for the development of robust classification methods using existing SAR instruments. [ABSTRACT FROM AUTHOR]

The Lee sigma filter was developed in 1983 based on the simple concept of two-sigma probability, and it was reasonably effective in speckle filtering. However, deficiencies were discovered in producing biased estimation and in blurring and depressing strong reflected targets. The advancement of synthetic aperture radar (SAR) technology with high-resolution data of large dimensions demands better and efficient speckle filtering algorithms. In this paper, we extend and improve the Lee sigma filter by eliminating these deficiencies. The bias problem is solved by redefining the sigma range based on the speckle probability density functions. To mitigate the problems of blurring and depressing strong reflective scatterers, a target signature preservation technique is developed. In addition, we incorporate the minimum-mean-square-error estimator for adaptive speckle reduction. Simulated SAR data are used to quantitatively evaluate the characteristics of this improved sigma filter and to validate its effectiveness. The proposed algorithm is applied to spaceborne and airborne SAR data to demonstrate its overall speckle filtering characteristics as compared with other algorithms. This improved sigma filter remains simple in concept and is computationally efficient but without the deficiencies of the original Lee sigma filter. [ABSTRACT FROM PUBLISHER]

Landsat data are now available for more than 30 years, providing the longest high-resolution record of Earth monitoring. This unprecedented time series of satellite imagery allows for extensive temporal observation of terrestrial processes such as land cover and land use change. However, despite this unique opportunity, most existing change detection techniques do not fully capitalize on this long time series. In this paper, a method that exploits both the temporal and spatial domains of time series satellite data to map land cover changes is presented. The time series of each pixel in the image is modeled with a combination of: 1) pixel-specific remotely sensed data; 2) neighboring pixels derived from ground observation data; and 3) time series transition probabilities. The spatial information is modeled with variograms and integrated using indicator kriging; time series transition probabilities are combined using an information-based cascade approach. This results in a map that is significantly more accurate in identifying when, where, and what land cover changes occurred. For the six images used in this paper, the prediction accuracy of the time series improves significantly, increasing from 31% to 61%, when both space and time are considered with the maximum likelihood. The consideration of spatial continuity also reduced unwanted speckles in the classified images, removing the need for any postprocessing. These results indicate that combining space and time domains significantly improves the accuracy of temporal change detection analyses and can produce high-quality time series land cover maps. [ABSTRACT FROM AUTHOR]