Search

Anomaly detection has been known to be a challenging problem due to the uncertainty of anomaly and the interference of noise. In this paper, we focus on anomaly detection in hyperspectral images (HSIs) and propose a novel detection algorithm based on spectral unmixing and dictionary-based low-rank decomposition. The innovation is threefold. First, due to the highly mixed nature of pixels in HSI data, instead of using the raw pixel directly for anomaly detection, the proposed algorithm applies spectral unmixing to obtain the abundance vectors and uses these vectors for anomaly detection. We show that the abundance vectors possess more distinctive features to identify anomaly from background. Second, to better represent the highly correlated background and the sparse anomaly, we construct a dictionary based on the mean shift clustering of the abundance vectors to improve both the discriminative and representative powers of the algorithm. Finally, a low-rank matrix decomposition method based on the constructed dictionary is proposed to encourage the coefficients of the dictionary, instead of the background itself, to be low rank, and the residual matrix to be sparse. Anomalies can then be extracted by summing up the columns of the residual matrix. The proposed algorithm is evaluated on both synthetic and real data sets. Experimental results show that the proposed approach constantly achieves high detection rate, while maintaining low false alarm rate regardless of the type of images tested. [ABSTRACT FROM AUTHOR]

This paper focuses on the endmember extraction (EE) technique for analyzing hyperspectral images. We first prove that the reconstruction errors (REs) and abundance anomalies (AAs) (abundances that fail to satisfy the abundance constraints) are effective in extracting undetected endmembers. Then, according to the spatial continuity of the endmember objects and differing from noise or outliers with a sparse distribution, the endmembers are assumed to be located at some salient areas in the RE and AA maps. A novel EE algorithm termed saliency-based endmember detection (SED) is proposed, where the visual saliency model is introduced to explore and analyze the spatial information that is contained in the AA and RE maps. Specifically, the AA and RE maps are regarded as the visual inputs, whereas the endmembers are treated as the visual stimuli. In SED, we assume that the pure pixel assumption holds. Based on the characteristics of the human visual system, the proposed method can not only extract endmembers in homogenous areas, but it can also highlight the small targets whose abundances may be spatially varied. In addition, since the spatial information is exploited in the reconstruction, the capability of the endmembers to represent the hyperspectral scene is automatically considered in the process of EE, and the detected endmembers are both accurate and reliable. The experimental results obtained on both simulated and real hyperspectral data confirm the merits and viability of the proposed algorithm. [ABSTRACT FROM AUTHOR]

Many hyperspectral anomaly detectors are designed based on the traditional Mahalanobis distance-based RX algorithm, which is usually considered as an inverse operation of the principal components analysis. Such detectors include the uniform target detector (UTD) algorithm, RX-UTD algorithm, and so on. However, the possibility of background statistical contamination caused by anomalies still exists. In order to alleviate this problem, in this paper, we propose a spectral matched filter (background-anomaly component projection and separation optimized filter) to minimize the average output energy of separate image components and the output values of the weighted background regular term for hyperspectral image anomaly detection, which could strengthen the separation between anomalies and backgrounds. By calculating the optimal solution to the background-anomaly component projection and separation function, we obtain the optimal projection, where we can effectively suppress the background while highlighting the anomalies. Proposed algorithm has the following research advantages: 1) it creates a novel collaborative component projection and robust background optimization function to separate the background and anomalies and 2) it analyzes the intrinsic statistical distribution of pixels and applies appropriate iterative shrinkage-thresholding algorithm to solve the $\ell _{1}$ -min problem. Experiments were conducted on three real hyperspectral data sets. The detection results demonstrate that the proposed algorithm is superior to other state-of-the-art anomaly detection algorithms. [ABSTRACT FROM AUTHOR]

Satellite remote-sensing data are essential for monitoring and quantifying water properties in the Great Lakes, providing useful monitoring and management tools for understanding water optical, biological, and ecological processes and phenomena. However, during the winter season, large parts of the Great Lakes are often covered by ice, which can cause significant uncertainties in satellite-measured water quality products. Although some developed radiance-based ice-detection algorithms for satellite ocean color data processing can eliminate most of the ice pixels in a region, there are still some significant errors due to misidentification of ice-contaminated pixels, particularly for the thin ice-covered regions. Therefore, it is necessary to improve the ice-detection methods for satellite ocean color data processing in the Great Lakes. In this paper, impacts of ice contamination on satellite-derived ocean color products in the Great Lakes are investigated, and a refined regional ice-detection algorithm which is based on the radiance spectra and normalized water-leaving radiance at the wavelength of 551 nm, nLw(551), is developed and assessed for satellite ocean color data processing in the Great Lakes. Results show that this proposed ice-detection method can reasonably identify ice-contaminated pixels, including those in very thin ice-covered regions, and provide accurate satellite ocean color products for the winter season in the Great Lakes. [ABSTRACT FROM PUBLISHER]

In this paper, we investigate the constant false-alarm rate (CFAR) property of the RX anomaly detector which is widely used for the analysis of hyperspectral data. The RX detector relies on an adaptive scheme where the mean vector and the covariance matrix of the background are locally estimated from the image pixels themselves. First, demeaning is accomplished by removing the estimated local background mean value, and then, the covariance matrix is estimated in a homogeneous neighborhood of each pixel. In principle, if the local mean is perfectly removed and the covariance matrix is estimated from background pixels sharing the same covariance matrix, the RX algorithm has the CFAR property, which is highly desirable in practical applications. The CFAR behavior of the algorithm also requires the spatial stationarity of the random noise affecting the hyperspectral image. In data collected by new-generation sensors, such an assumption is not valid because photon noise contribution, which depends on the spatially varying signal level, is not negligible. This has motivated us to analyze the behavior of the RX algorithm with respect to the CFAR property in data affected by signal-dependent (SD) noise. In this paper, we show both theoretically and experimentally that the SD noise is one of the causes of the non-CFAR behavior of the RX detector that we have experienced in many practical situations. We propose a strategy to enhance the robustness of the anomaly detection scheme with respect to the CFAR property based on an adaptive nonlinear transform aimed at reducing the dependence of the noise on the signal level. Experiments on simulated data and real data collected by a new hyperspectral camera are also presented and discussed. [ABSTRACT FROM AUTHOR]

In this paper, a graph-based nonlocal total variation method is proposed for unsupervised classification of hyperspectral images (HSI). The variational problem is solved by the primal-dual hybrid gradient algorithm. By squaring the labeling function and using a stable simplex clustering routine, an unsupervised clustering method with random initialization can be implemented. The effectiveness of this proposed algorithm is illustrated on both synthetic and real-world HSI, and numerical results show that the proposed algorithm outperforms other standard unsupervised clustering methods, such as spherical $K$ -means, nonnegative matrix factorization, and the graph-based Merriman–Bence–Osher scheme. [ABSTRACT FROM PUBLISHER]

Investigations in data quality and uncertainty modeling are becoming key topics in geoinformation science. This paper models a collection of outcomes from a standard segmentation algorithm as a random set. It quantifies extensional uncertainties of extracted objects using statistical characteristics of random sets. The approach is applied to a synthetic data set and vegetation patches in the Poyang Lake area in China. These patches are of interest as they have both sharp and vague boundaries. Results show that random sets provide useful spatial information on uncertainties using their basic parameters like the mean, level sets, and variance. The number of iterations to achieve a stable covering function and the sum of the variances are good indicators of boundary sharpness. The coefficient of variation has a positive relation with the degree of uncertainty. An asymmetry ratio reflects the uneven gradual changes along different directions where broad boundaries exist. This paper shows that several characteristics of extensional uncertainty of segmented objects can be quantified numerically and spatially by random sets. [ABSTRACT FROM AUTHOR]

This paper deals with subspace-based target detection in hyperspectral images. Specifically, it focuses on a general detection scheme where, first, background is suppressed through orthogonal-subspace projection and then target detection is accomplished. An adequate estimation of the background subspace is essential to a successful outcome. The background subspace has been typically estimated globally. However, global approaches may be ineffective for small-target-detection applications since they tend to overestimate the background interference affecting a given target. This may result in a low target residual energy after background suppression that is detrimental to detection performance. In this paper, we propose a novel and fully automatic algorithm for local background-subspace estimation (LBSE). Local background has typically a lower inherent complexity than that of global background. By estimating the background subspace over a local neighborhood of the test pixel, the resulting background-subspace dimension is expected to be low, thus resulting in a higher target residual energy after suppression which benefits the detection performance. Specifically, the proposed LBSE acts on a per-pixel basis, thus adaptively tailoring the estimated basis to the local complexity of background. Both simulated and real hyperspectral data are employed to investigate the detection-performance improvements offered by LBSE with respect to both global and local methodologies previously presented. [ABSTRACT FROM AUTHOR]

Hyperspectral images exhibit strong dependencies across spatial and spectral neighbors, which have been proved to be very useful for hyperspectral image classification. The recently defined conditional random field (CRF) can effectively model and use the dependencies for classification of hyperspectral images in a unified probabilistic framework. However, in order to be computationally tractable, the usual CRFs are limited to incorporate only pairwise potentials. Thus, the usual CRFs can capture only pairwise interactions and neglect higher order dependencies, which are potentially useful high-level properties particularly for the classification of hyperspectral image consisting of complex components. This paper overcomes this limitation by developing hyperspectral image classification algorithm based on a CRF with sparse higher order potentials, which are specially designed to incorporate complex characteristics of hyperspectral images. To efficiently implement the CRF model at training step, this paper develops an efficient local method under the piecewise training framework, while at inference step, this proposes a simple strategy to combine the piecewisely trained model to overcome the possible over-counting problems. Moreover, the combined model with the specially defined potentials can be efficiently inferred by graph cut method. Experiments on the real-world data attest to the accuracy, effectiveness, and efficiency of the proposed model on modeling and classifying hyperspectral images. [ABSTRACT FROM AUTHOR]

This paper presents two semisupervised one-class support vector machine (OC-SVM) classifiers for remote sensing applications. In one-class image classification, one tries to detect pixels belonging to one of the classes in the image and reject the others. When few labeled pixels of only one class are available, obtaining a reliable classifier is a difficult task. In the particular case of SVM-based classifiers, this task is even harder because the free parameters of the model need to be finely adjusted, but no clear criterion can be adopted. In order to improve the OC-SVM classifier accuracy and alleviate the problem of free-parameter selection, the information provided by unlabeled samples present in the scene can be used. In this paper, we present two state-of-the-art algorithms for semisupervised one-class classification for remote sensing classification problems. The first proposed algorithm is based on modifying the OC-SVM kernel by modeling the data marginal distribution with the graph Laplacian built with both labeled and unlabeled samples. The second one is based on a simple modification of the standard SVM cost function which penalizes more the errors made when classifying samples of the target class. The good performance of the proposed methods is illustrated in four challenging remote sensing image classification scenarios where the goal is to detect one of the classes present on the scene. In particular, we present results for multisource urban monitoring, hyperspectral crop detection, multispectral cloud screening, and change-detection problems. Experimental results show the suitability of the proposed techniques, particularly in cases with few or poorly representative labeled samples. [ABSTRACT FROM AUTHOR]

This paper deals with the problem of signal subspace estimation for dimensionality reduction (DR) in hyperspectral images in the presence of rare pixels, i.e., pixels that are scarcely represented in the image and containing spectral components that are linearly independent of the background. Most of the classical methods proposed in the literature are based on the analysis second-order statistics (SOS), which are weakly influenced by the rare signals. Therefore, such techniques estimate the signal subspace addressing mostly the background and ignoring the presence of rare pixels. This may reduce the target/background spectral contrast, thus decreasing the detection performance when DR is adopted as preprocessing task in small-target detection applications. In this paper, a new robust algorithm, namely, robust signal subspace estimation (RSSE), is developed, which preserves both abundant and rare signal components. It combines the analysis of SOS and a recent approach based on the analysis of the l2∞ norm. The novel contribution of this paper is twofold. First, the RSSE algorithm is presented, which includes a new iterative procedure to derive the signal subspace and an original statistical method to estimate the data dimensionality. Second, an ad simulation strategy is proposed to assess the performance of signal subspace estimation methods in the presence of rare signal components. The procedure is adopted to compare the RSSE algorithm with a classical technique based on the analysis of SOS. The results obtained by applying the two methods on a real Airborne Visible Infrared Imaging Spectrometer hyperspectral image also presented and discussed. [ABSTRACT FROM AUTHOR]

In blind hyperspectral unmixing (HU), the pure-pixel assumption is well known to be powerful in enabling simple and effective blind HU solutions. However, the pure-pixel assumption is not always satisfied in an exact sense, especially for scenarios where pixels are heavily mixed. In the no-pure-pixel case, a good blind HU approach to consider is the minimum volume enclosing simplex (MVES). Empirical experience has suggested that MVES algorithms can perform well without pure pixels, although it was not totally clear why this is true from a theoretical viewpoint. This paper aims to address the latter issue. We develop an analysis framework wherein the perfect endmember identifiability of MVES is studied under the noiseless case. We prove that MVES is indeed robust against lack of pure pixels, as long as the pixels do not get too heavily mixed and too asymmetrically spread. The theoretical results are supported by numerical simulation results. [ABSTRACT FROM PUBLISHER]

In this paper, we propose a superpixel-level sparse representation classification framework with multitask learning for hyperspectral imagery. The proposed algorithm exploits the class-level sparsity prior for multiple-feature fusion, and the correlation and distinctiveness of pixels in a spatial local region. Compared with some of the state-of-the-art hyperspectral classifiers, the superiority of the multiple-feature combination, the spatial prior utilization, and the computational complexity are maintained at the same time in the proposed method. The proposed classification algorithm was tested on three hyperspectral images. The experimental results suggest that the proposed algorithm performs better than the other sparse (collaborative) representation-based algorithms and some popular hyperspectral multiple-feature classifiers. [ABSTRACT FROM PUBLISHER]

This paper describes the principles and implementation of an algorithm for the classification of hyperspectral remote sensing images. The proposed approach is novel and can be included within the category of the spectral–spatial classification algorithms. The elements of novelty of the algorithm are as follows: 1) the implementation of two classifiers that work iteratively, each one exploiting the decision of the other to improve the training phase, and 2) the use of relational features based on the current labeling and on the spatial structure of the image. The two classifiers are fed with the spectral features and with the spatial features, respectively. The spatial features are built using the relative abundance of each class in a neighborhood of the pixel (homogeneity index), where the neighborhood is properly defined. An important contribution to the success of the method is the adoption of a multiclass classifier, the multinomial logistic regression, and a proper use of the posterior probabilities to infer the class labeling and build the relational data. The results of the two classifiers are eventually combined by means of an ensemble decision. The algorithm has been successfully tested on three standard hyperspectral images taken from the Airborne Visible–Infrared Imaging Spectrometer and ROSIS airborne sensors and compared with classification algorithms recently proposed in the literature. [ABSTRACT FROM PUBLISHER]

This paper presents a solution to solve the car detection and counting problem in images acquired by means of unmanned aerial vehicles (UAVs). UAV images are characterized by a very high spatial resolution (order of few centimeters), and consequently by an extremely high level of details which calls for appropriate automatic analysis methods. The proposed method starts with a screening step of asphalted zones in order to restrict the areas where to detect cars and thus to reduce false alarms. Then, it performs a feature extraction process based on scalar invariant feature transform thanks to which a set of keypoints is identified in the considered image and opportunely described. Successively, it discriminates between keypoints assigned to cars and all the others, by means of a support vector machine classifier. The last step of our method is focused on the grouping of the keypoints belonging to the same car in order to get a “one keypoint-one car” relationship. Finally, the number of cars present in the scene is given by the number of final keypoints identified. The experimental results obtained on a real UAV scene characterized by a spatial resolution of 2 cm show that the proposed method exhibits a promising car counting accuracy. [ABSTRACT FROM PUBLISHER]

In previous works, a general framework to exploit polarimetric diversity to optimize the results of persistent scatterers interferometry (PSI) was presented, but tested only with dual-pol data. In this paper, the performance of these algorithms is assessed using fully polarimetric data, acquired by the Radarsat-2 satellite over the urban area of Barcelona, Spain. In addition, two new highly efficient polarimetric optimization methods, mean intensity polarimetric optimization and joint diagonalization-based polarimetric optimization, are introduced and evaluated. Given the variety of dual-pol configurations provided by current polarimetric satellites, such as TerraSAR-X and Radarsat-2, and the upcoming launch of Sentinel-1, ALOS-2, and Radarsat Constellation Mission, a study has been also carried out to determine the best performing dual-pol configurations for polarimetric PSI. Subsidence maps of the area of study are computed for single-pol, dual-pol, and full-pol data, which show the increase in pixel density with valid deformation results as more polarimetric information is made available. In particular, for full-pol data we get an increase of up to 2.5 times more pixels for coherence-based PSI techniques (degraded resolution), and over four times more for amplitude-based approaches (full resolution), in comparison with single-pol data. Both higher density and quality of pixels yield better results in terms of coverage and accuracy. [ABSTRACT FROM PUBLISHER]

Filling dead pixels or removing uninteresting objects is often desired in the applications of remotely sensed images. In this paper, an effective image inpainting technology is presented to solve this task, based on multichannel nonlocal total variation. The proposed approach takes advantage of a nonlocal method, which has a superior performance in dealing with textured images and reconstructing large-scale areas. Furthermore, it makes use of the multichannel data of remotely sensed images to achieve spectral coherence for the reconstruction result. To optimize the proposed variation model, a Bregmanized-operator-splitting algorithm is employed. The proposed inpainting algorithm was tested on simulated and real images. The experimental results verify the efficacy of this algorithm. [ABSTRACT FROM AUTHOR]

Hyperspectral unmixing is one of the most important techniques in analyzing hyperspectral images, which decomposes a mixed pixel into a collection of constituent materials weighted by their proportions. Recently, many sparse nonnegative matrix factorization (NMF) algorithms have achieved advanced performance for hyperspectral unmixing because they overcome the difficulty of absence of pure pixels and sufficiently utilize the sparse characteristic of the data. However, most existing sparse NMF algorithms for hyperspectral unmixing only consider the Euclidean structure of the hyperspectral data space. In fact, hyperspectral data are more likely to lie on a low-dimensional submanifold embedded in the high-dimensional ambient space. Thus, it is necessary to consider the intrinsic manifold structure for hyperspectral unmixing. In order to exploit the latent manifold structure of the data during the decomposition, manifold regularization is incorporated into sparsity-constrained NMF for unmixing in this paper. Since the additional manifold regularization term can keep the close link between the original image and the material abundance maps, the proposed approach leads to a more desired unmixing performance. The experimental results on synthetic and real hyperspectral data both illustrate the superiority of the proposed method compared with other state-of-the-art approaches. [ABSTRACT FROM PUBLISHER]

Hyperspectral endmember extraction is a process to estimate endmember signatures from the hyperspectral observations, in an attempt to study the underlying mineral composition of a landscape. However, estimating the number of endmembers, which is usually assumed to be known a priori in most endmember estimation algorithms (EEAs), still remains a challenging task. In this paper, assuming hyperspectral linear mixing model, we propose a hyperspectral data geometry-based approach for estimating the number of endmembers by utilizing successive endmember estimation strategy of an EEA. The approach is fulfilled by two novel algorithms, namely geometry-based estimation of number of endmembers—convex hull (GENE-CH) algorithm and affine hull (GENE-AH) algorithm. The GENE-CH and GENE-AH algorithms are based on the fact that all the observed pixel vectors lie in the convex hull and affine hull of the endmember signatures, respectively. The proposed GENE algorithms estimate the number of endmembers by using the Neyman–Pearson hypothesis testing over the endmember estimates provided by a successive EEA until the estimate of the number of endmembers is obtained. Since the estimation accuracies of the proposed GENE algorithms depend on the performance of the EEA used, a reliable, reproducible, and successive EEA, called p-norm-based pure pixel identification (TRI-P) algorithm is then proposed. The performance of the proposed TRI-P algorithm, and the estimation accuracies of the GENE algorithms are demonstrated through Monte Carlo simulations. Finally, the proposed GENE and TRI-P algorithms are applied to real AVIRIS hyperspectral data obtained over the Cuprite mining site, Nevada, and some conclusions and future directions are provided. [ABSTRACT FROM PUBLISHER]

Denoising of hyperspectral imagery in the domain of imaging spectroscopy by conditional random fields (CRFs) is addressed in this work. For denoising of hyperspectral imagery, the strong dependencies across spatial and spectral neighbors have been proved to be very useful. Many available hyperspectral image denoising algorithms adopt multidimensional tools to deal with the problems and thus naturally focus on the use of the spectral dependencies. However, few of them were specifically designed to use the spatial dependencies. In this paper, we propose a multiple-spectral-band CRF (MSB-CRF) to simultaneously model and use the spatial and spectral dependencies in a unified probabilistic framework. Furthermore, under the proposed MSB-CRF framework, we develop two hyperspectral image denoising algorithms, which, thanks to the incorporated spatial and spectral dependencies, can significantly remove the noise, while maintaining the important image details. The experiments are conducted in both simulated and real noisy conditions to test the proposed denoising algorithms, which are shown to outperform the popular denoising methods described in the previous literatures. [ABSTRACT FROM PUBLISHER]

The amount of noise included in a hyperspectral image limits its application and has a negative impact on hyperspectral image classification, unmixing, target detection, and so on. In hyperspectral images, because the noise intensity in different bands is different, to better suppress the noise in the high-noise-intensity bands and preserve the detailed information in the low-noise-intensity bands, the denoising strength should be adaptively adjusted with the noise intensity in the different bands. Meanwhile, in the same band, there exist different spatial property regions, such as homogeneous regions and edge or texture regions; to better reduce the noise in the homogeneous regions and preserve the edge and texture information, the denoising strength applied to pixels in different spatial property regions should also be different. Therefore, in this paper, we propose a hyperspectral image denoising algorithm employing a spectral–spatial adaptive total variation (TV) model, in which the spectral noise differences and spatial information differences are both considered in the process of noise reduction. To reduce the computational load in the denoising process, the split Bregman iteration algorithm is employed to optimize the spectral–spatial hyperspectral TV model and accelerate the speed of hyperspectral image denoising. A number of experiments illustrate that the proposed approach can satisfactorily realize the spectral–spatial adaptive mechanism in the denoising process, and superior denoising results are produced. [ABSTRACT FROM PUBLISHER]

A variety of competing algorithms exist for segmentation of both single-channel and multichannel synthetic aperture radar (SAR) images. Among the most successful of these algorithms is the approach presented by Stewart This algorithm defines a cost which is a weighted sum of a likelihood term that estimates the statistical likelihood of the membership of pixels to neighboring segments and a shape term that is intended to provide a smoothing constraint on segment boundaries. The shape term in the original implementation of the Stewart algorithm was rather rudimentary, and in this paper, we explore the performance of a shape term based on Sethian–Osher curvature-flow theory. We demonstrate the performance of the refined curvature-cost (CC) shape term on a set of simulated images as well as an ASAR scene. We assess the segmentation performance using a hybridized shape metric and on the number of segments produced. We find that the CC shape term significantly improves the performance of the Stewart segmentation algorithm, particularly for high-contrast edges. In spite of this success, we argue that further improvements to the algorithm will be difficult due to the architecture of the system. [ABSTRACT FROM PUBLISHER]

In the late 1990s, Winter proposed an endmember extraction belief that has much impact on endmember extraction techniques in hyperspectral remote sensing. The idea is to find a maximum-volume simplex whose vertices are drawn from the pixel vectors. Winter's belief has stimulated much interest, resulting in many different variations of pixel search algorithms, widely known as N-FINDR, being proposed. In this paper, we take a continuous optimization perspective to revisit Winter's belief, where the aim is to provide an alternative framework of formulating and understanding Winter's belief in a systematic manner. We first prove that, fundamentally, the existence of pure pixels is not only sufficient for the Winter problem to perfectly identify the ground-truth endmembers but also necessary. Then, under the umbrella of the Winter problem, we derive two methods using two different optimization strategies. One is by alternating optimization. The resulting algorithm turns out to be an N-FINDR variant, but, with the proposed formulation, we can pin down some of its convergence characteristics. Another is by successive optimization; interestingly, the resulting algorithm is found to exhibit some similarity to vertex component analysis. Hence, the framework provides linkage and alternative interpretations to these existing algorithms. Furthermore, we propose a robust worst case generalization of the Winter problem for accounting for perturbed pixel effects in the noisy scenario. An algorithm combining alternating optimization and projected subgradients is devised to deal with the problem. We use both simulations and real data experiments to demonstrate the viability and merits of the proposed algorithms. [ABSTRACT FROM AUTHOR]

This paper presents an automatic image segmentation method for polarimetric synthetic aperture radar data. It utilizes the full polarimetric information and incorporates texture by modeling with a non-Gaussian distribution for the complex scattering coefficients. The modeling is based upon the well-known product model, with a Gamma-distributed texture parameter leading to the K-Wishart model for the covariance matrix. The automatic clustering is achieved through a finite mixture model estimated with a modified expectation maximization algorithm. We include an additional goodness-of-fit test stage that allows for splitting and merging of clusters. This not only improves the model fit of the clusters, but also dynamically selects the appropriate number of clusters. The resulting image segmentation depicts the statistically significant clusters within the image. A key feature is that the degree of sub-sampling of the input image will affect the detail level of the clustering, revealing only the major classes or a variable level of detail. Real-world examples are shown to demonstrate the technique. [ABSTRACT FROM AUTHOR]

In this paper, a new algorithm for striping noise reduction in hyperspectral images is proposed. The new algorithm exploits the orthogonal subspace approach to estimate the striping component and to remove it from the image, preserving the useful signal. The algorithm does not introduce artifacts in the data and also takes into account the dependence on the signal intensity of the striping component. The effectiveness of the algorithm in reducing striping noise is experimentally demonstrated on real data acquired both by airborne and satellite hyperspectral sensors. [ABSTRACT FROM AUTHOR]

Target detection is a key issue in processing hyperspectral images (HSIs). Spectral-identification-based algorithms are sensitive to spectral variability and noise in acquisition. In most cases, both the target spatial distributions and the spectral signatures are unknown, so each pixel is separately tested and appears as a target when it significantly differs from the background. In this paper, we propose two algorithms to improve the signal-to-noise ratio (SNR) of hyperspectral data, leading to detectors that are robust to noise. These algorithms consist in integrating adaptive spatial/spectral filtering into the adaptive matched filter and adaptive coherence estimator. Considering the HSIs as tensor data, our approach introduces a data representation involving multidimensional processing. It combines the advantages of spatial and spectral information using an alternating least square algorithm. To estimate the signal subspace dimension in each spatial mode, we extend the Akaike information criterion, and we develop an iterative algorithm for spectral-mode rank estimation. We demonstrate the interest of integrating the quadtree decomposition to perform an adaptive 3-D filtering and thereby preserve the local image characteristics. This leads to a significant improvement in terms of denoised tensor SNR and, consequently, in terms of detection probability. The performance of our method is exemplified using simulated and real-world HYperspectral Digital Imagery Collection Experiment images. [ABSTRACT FROM AUTHOR]

This paper addresses the remote sensing image pan-sharpening problem from the perspective of compressed sensing (CS) theory which ensures that with the sparsity regularization, a compressible signal can be correctly recovered from the global linear sampled data. First, the degradation model from a high- to low-resolution multispectral (MS) image and high-resolution panchromatic (PAN) image is constructed as a linear sampling process which is formulated as a matrix. Then, the model matrix is considered as the measurement matrix in CS, so pan-sharpening is converted into signal restoration problem with sparsity regularization. Finally, the basis pursuit (BP) algorithm is used to resolve the restoration problem, which can recover the high-resolution MS image effectively. The QuickBird and IKONOS satellite images are used to test the proposed method. The experimental results show that the proposed method can well preserve spectral and spatial details of the source images. The pan-sharpened high-resolution MS image by the proposed method is competitive or even superior to those images fused by other well-known methods. [ABSTRACT FROM AUTHOR]

Nonnegative matrix factorization (NMF) has been recently applied to solve the hyperspectral unmixing problem because it ensures nonnegativity and needs no assumption for the presence of pure pixels. However, the algorithm has a large amount of local minima due to the obvious nonconvexity of the objective function. In order to improve its performance, auxiliary constraints can be introduced into the algorithm. In this paper, we propose a new approach named abundance separation and smoothness constrained NMF by introducing two constraints, namely, abundance separation and smoothness, into the NMF algorithm. These constraints are based on two properties of hyperspectral imagery. First, usually, every ground object presents dominance in a specific region of the entire image scene and the correlation is weak between different endmembers. Second, moving through various regions, ground objects usually vary slowly and abrupt changes rarely appear. We also propose a learning algorithm to further improve the performance of our method, from which the auxiliary constraints are removed at an appropriate time. The proposed algorithm retains all the advantages of NMF and effectively overcomes the shortcoming of local minima at the same time. Experimental results based on synthetic and real hyperspectral data show the superiority of the proposed algorithm with respect to other state-of-the-art approaches. [ABSTRACT FROM AUTHOR]

This paper proposes a new method that solves the problem of inverse synthetic aperture radar image cross-range scaling by estimating the rotational velocity (RV) using the expansion–rotation–scale relationship between two range–Doppler (RD) images. This method is composed of three steps. The first step is preprocessing to construct 2-D Fourier transform images and initial polar-mapped images. In this step, two RD images are 2-D Fourier transformed to avoid the necessity of finding the rotation center; then, the transformed images are polar mapped with identical polar grids to convert rotation into translation in the \theta-direction only. The second step is a coarse search that finds the angular shift that provides the maximum correlation between two polar images. The angular shift found is used as the initial relative rotation angle (RA), and the initial relative scaling factor (RSF) is calculated using the RV which is equal to the angular shift divided by the time delay between the images. The third step is the optimization of the relative RA using the Nelder–Mead approach, with the RSF updated using the relative RA derived during each iteration. In simulations using a Mig-25 aircraft, composed of ideal point scatterers, and the measured data from a Boeing 747-400, the targets were properly rescaled in the range–cross-range domain due to the accurate estimation of the RV. [ABSTRACT FROM AUTHOR]

In this paper, heterogeneous clutter models are used to describe polarimetric synthetic aperture radar (PolSAR) data. The KummerU distribution is introduced to model the PolSAR clutter. Then, a detailed analysis is carried out to evaluate the potential of this new multivariate distribution. It is implemented in a hierarchical maximum likelihood segmentation algorithm. The segmentation results are shown on both synthetic and high-resolution PolSAR data at the X- and L-bands. Finally, some methods are examined to determine automatically the “optimal” number of segments in the final partition. [ABSTRACT FROM AUTHOR]

Multibaseline phase unwrapping is a critical processing procedure of multibaseline synthetic aperture radar interferometry (InSAR). It has an advantage over single-baseline phase unwrapping in discontinuous-terrain-height estimation. In this paper, the pixels' combination information of multiple InSAR interferograms with different baseline lengths is deeply investigated, and the term ambiguity vector is proposed to represent a pixel's ambiguity number of multiple interferograms. It is also revealed that pixels with the same ambiguity vector have an exclusive pattern among them. A fast cluster-analysis (CA)-based method is proposed for multibaseline phase unwrapping. In this method, all pixels are first clustered into different groups according to their patterns, and then, information of the cluster center is used to unwrap the phases of pixels group by group. Simulation results are shown to verify the effectiveness, efficiency, and noise robustness of CA multibaseline phase-unwrapping method. [ABSTRACT FROM AUTHOR]

In this paper, an extended COvariance Matching Estimation Techniques–Extend Invariance Principle (COMET-EXIP) method is proposed to estimate interferometric synthetic aperture radar or interferometric synthetic aperture sonar (InSAS) interferometric phase in the presence of large coregistration errors, even up to one pixel. First, the extended COMET-EXIP method is presented for the application of joint-pixel-model-based interferogram estimation, through choosing a novel “unstructured model” in terms of the parameters to be estimated and decoupling the interesting parameters from the uninteresting “nuisance parameters.” Then, a fast algorithm of COMET-EXIP is proposed for the interferometric phase estimation. Finally, the ambiguity problem of the COMET-EXIP method is solved without introducing performance degradation. The simulated data and real data from the trial InSAS and X-SAR are used to verify the validity of the method. The results show that the method is robust for a wide range of signal-to-noise ratio and has a good performance on both fringe preserving and noise suppressing. In addition, the same computational speed level of the proposed method as that of the pivoting mean filtering is very attractive. [ABSTRACT FROM AUTHOR]

Superpixels are homogeneous image regions comprised of multiple contiguous pixels. Superpixel representations can reduce noise in hyperspectral images by exploiting the spatial contiguity of scene features. This paper combines superpixels with endmember extraction to produce concise mineralogical summaries that assist in browsing large image catalogs. First, a graph-based agglomerative algorithm oversegments the image. We then use segments' mean spectra as input to existing statistical endmember detection algorithms such as sequential maximum angle convex cone (SMACC) and N-FINDR. Experiments compare automatically detected endmembers to target minerals in an Airborne Visible/Infrared Imaging Spectrometer (AVIRIS) scene of Cuprite, Nevada. We also consider a planetary science data set from the Compact Reconnaissance Imaging Spectrometer (CRISM) instrument that benefits from spatial averaging due to higher noise. In both cases, superpixel representations significantly reduce the computational complexity of later processing while improving endmembers' match to the target spectra. [ABSTRACT FROM AUTHOR]

The linear mixture model is a convenient way to describe image pixels as a linear combination of pure spectra – termed endmembers. The fractional contribution from each endmember is calculated through inversion of the linear model. Despite the simplicity of the model, a nonnegativity constraint that is imposed on the fractions leads to an unmixing problem for which it is hard to find a closed analytical solution. Current solutions to this problem involve iterative algorithms, which are computationally intensive and not appropriate for unmixing large number of pixels. This paper presents an algorithm to build fuzzy membership functions that are equivalent to the least square solution of the fully constrained linear spectral unmixing problem. The efficiency and effectiveness of the proposed solution is demonstrated using both simulated and real data. [ABSTRACT FROM AUTHOR]

The simulation of remote-sensing hyperspectral images is a useful tool for a variety of tasks such as the design of systems, the understanding of the image formation process, and the development and validation of data processing algorithms. The lack of ground truth and the incomplete knowledge of the Martian environment make simulation studies of Mars hyperspectral images a useful tool for automated analysis of Mars data. Hyperspectral near-infrared scenes of mineral mixtures have been simulated to analyze the contributions of surface minerals, atmosphere, and sensor noise on images of Mars. Modeling the remote-sensing process creates a means for the independent analysis of the influence of the environment and instruments on the detection accuracy of the surface composition (e.g., the scene endmembers). The end-to-end model builds surface reflectance scenes based on laboratory sample spectra, creates atmospheric effects using radiative transfer routines, simulates the instrument response function using CRISM data files, and adds instrument noise from thermal and other sources. The purpose of this paper is to understand the hyperspectral remote-sensing process to eventually enable the elevated detection accuracy of minerals on the surface of Mars. The viability of a linear approximation of the complete model is also investigated. The approximation is compared to the complete model in an image classification task. [ABSTRACT FROM AUTHOR]

To obtain depth-from-stereo imagery, it is traditionally required that the baseline separation between images (or the base-to-height ratio) be very large in order to ensure the largest image disparity range for effective measurement. Typically, a B/H ratio in the range of 0.6–1 is preferred. As a consequence, most existing stereo-matching algorithms are designed to measure disparities reliably with only integer-pixel precision. However, wide baselines may increase the possibility of occlusion occurring between highly contrasting relief, imposing a serious problem to digital elevation model (DEM) generation in urban and highly dissected mountainous areas. A narrow-baseline stereo configuration can alleviate the problem significantly but requires very precise measurements of disparity at subpixel levels. In this paper, we demonstrate a stereo-matching algorithm, based upon the robust phase correlation method, that is capable of directly measuring disparities up to 1/50th pixel accuracy and precision. The algorithm enables complete and dense surface shape information to be retrieved from images with unconventionally low B/H ratios (e.g., less than 0.01), potentially allowing DEM generation from images that would otherwise not be deemed suitable for the purpose. [ABSTRACT FROM AUTHOR]

Spectral mixture analysis (SMA) has been widely utilized to address the mixed-pixel problem in the quantitative analysis of hyperspectral remote sensing images, in which endmember extraction (EE) plays an extremely important role. In this paper, a novel algorithm is proposed to integrate both spectral similarity and spatial context for EE. The spatial context is exploited from two aspects. At first, initial endmember candidates are identified by determining the spatial purity (SP) of pixels in their spatial neighborhoods (SNs). Several SP measurements are investigated at both intensity level and feature level. In order to alleviate local spectra variability, the average of the pixels in pure SNs are voted as endmember candidates. Then, the spatial connectivity is utilized to merge spatially related endmember candidates by finding connection paths in a graph so that the number of endmember candidates is further reduced, which results in computational efficiency and better performance in SMA by alleviating global spectral variability. Experimental results on both synthetic and real hyperspectral images demonstrate that the proposed SP based EE (SPEE) algorithm outperforms the other popular EE algorithms. It is also observed that feature-level SP measurements are more distinguishable than intensity-level SP measurements to discriminate pure SNs from mixed SNs. [ABSTRACT FROM AUTHOR]

Whereas overhead infrared imagery shows promise for detecting buried landmines, detection algorithms must deal with the daunting challenge of distinguishing between landmines and clutter objects which frequently possess similar spatial and spectral characteristics to landmines. However, groups of clutter features are rarely related spatially in the same way that groups of mines are related. For this reason, the recognition of minefield patterns in overhead landmine imagery can be useful to the detection of mines in minefields. In this paper, we present a simple method for detecting grid patterns in imagery, discuss means by which the method may be extended to a more general category of patterns, provide a method for the automated prediction of the locations of undetected mines based upon the observed pattern, and finally we discuss applications. Examples are provided using longwave infrared hyperspectral imagery. [ABSTRACT FROM AUTHOR]

Maximization of mutual information (MI) (MMI) has a wide application in multimodal image registration. The conventional MMI algorithm is based on the statistics derived from image sampling and does not use spatial features. In this paper, we proposed an improved MMI registration algorithm that combines the spatial information through a feature-based selection mechanism. A Harris corner filter is chosen to classify the pixels. The classification results are used to focus sample selection, joint entropy estimation, and weighted MI information calculation. Although features play an important role, conventional feature matching is not used, which removes an important challenge to the use of feature-based information. The advantages of the proposed algorithm are improved robustness by incorporation of spatial features and significant reduction in computing time for registration. Experiments with challenging synthetic images and multimodal airborne infrared images are provided to demonstrate the improvement. [ABSTRACT FROM AUTHOR]

This paper considers the problem of unsupervised spectral unmixing for hyperspectral image analysis. Each observed pixel is assumed to be a noisy linear mixture of pure material spectra, namely, endmembers. The mixing coefficients, usually called abundances, are constrained to positive and summed to unity. The proposed unmixing approach is based on the nonnegative matrix factorization (NMF) framework, which considers the physical constraints of the problem, including the positivity of the endmember spectra and abundances. However, the basic NMF formulation has degenerated solutions and suffers from nonconvexity limitations. We consider here a regularization function, called dispersion, which favors the solution such that the endmember spectra have minimum variances. Such a solution encourages the recovered spectra to be flat, preserving the possible spectral singularities (peaks and sharp variations). The regularized criterion is minimized with a projected gradient (PG) scheme, and we propose a new step-size estimation technique to fasten the PG convergence. The derived algorithm is called MiniDisCo, for minimum dispersion constrained NMF. We experimentally compare MiniDisCo with the recently proposed algorithm. It is shown to be particularly robust to the presence of flat spectra, to a possible a priori overestimation of the number of endmembers, or if the amount of observed spectral pixels is low. In addition, experiments show that the considered regularization correctly overcomes the degeneracy and nonconvexity problems, leading to satisfactory unmixing accuracy. We include a comparative analysis of a real-world scene. [ABSTRACT FROM AUTHOR]

In this paper, we propose two active learning algorithms for semiautomatic definition of training samples in remote sensing image classification. Based on predefined heuristics, the classifier ranks the unlabeled pixels and automatically chooses those that are considered the most valuable for its improvement. Once the pixels have been selected, the analyst labels them manually and the process is iterated. Starting with a small and nonoptimal training set, the model itself builds the optimal set of samples which minimizes the classification error. We have applied the proposed algorithms to a variety of remote sensing data, including very high resolution and hyperspectral images, using support vector, machines. Experimental results confirm the consistency of the methods. The required number of training samples can be reduced to 10% using the methods proposed, reaching the same level of accuracy as larger data sets. A comparison with a state-of-the-art active learning method, margin sampling, provided, highlighting advantages of the methods proposed. The effect of spatial resolution and separability of the classes on the quality of the selection of pixels is also discussed. [ABSTRACT FROM AUTHOR]

This paper forwards an advanced subpixel mapping algorithm to provide detailed information on the spatial distribution of land covers within a mixed pixel. This is achieved utilizing the area proportions of the endmember components a mixed pixel and its neighboring pixels. Fraction values, obtained through soft classification, are used to calculate the area ratios of the endmember components of the mixed pixel and neighboring pixels. After testing the algorithm with both artificial and synthetic images, the performance of the algorithm can be evaluated as being computationally efficient and accurate for obtaining comprehensive information on the spatial distribution of land covers. [ABSTRACT FROM AUTHOR]

Many endmember extraction algorithms (EEAs) have been developed to find endmembers that are assumed to be pure signatures in hyperspectral data. However, two issues arising in EEAs have not been addressed: one is the knowledge of the number of endmembers that must be provided a priori, and the other is the initialization of EEAs, where most EEAs initialize their endmember-searching processes by using randomly generated endmembers, which generally result in inconsistent final selected endmembers. Unfortunately, there has been no previous work reported on how to address these two issues, i.e., how to select a set of appropriate initial endmembers and how to determine the number of endmembers p. This paper takes up these two issues and describes two-stage processes to improve EEAs. First, a recently developed concept of virtual dimensionality (VD) is used to determine how many endmembers are needed to be generated for an EEA. Experiments show that the VD is an adequate mea- sure for estimating p. Second, since EEAs are sensitive to initial endmembers, a properly selected set of initial endmembers can make significant improvements on the searching process. In doing so, a new concept of endmember initialization algorithm (EIA) is thus proposed, and four different algorithms are suggested for this purpose. It is surprisingly found that many EIA-generated initial endmembers turn out to be the final desired endmembers. A further objective is to demonstrate that EEAs implemented in conjunction with EIA-generated initial endmembers can significantly reduce the number of endmember replacements as well as the computing time during endmember search. [ABSTRACT FROM AUTHOR]

In this paper, the problem of change detection from synthetic aperture radar (SAR) images is addressed. Feature-level change-detection algorithms are still in their preliminary design stage. Indeed, while pixel-based approaches are already implemented into existing, commercial software, this is not the case for feature comparison approaches. Here, the authors propose a joint use of both approaches. The approach is based on the extraction and comparison of linear features from multiple SAR images, to confirm pixel-based changes. Though simple, the methodology proves to be effective, irrespectively of misregistration errors due to reprojection problems or difference in the sensor's viewing geometry, which are common in multitemporal SAR images. The procedure is validated through synthetic examples, but also two real change-detection situations, using airborne and satellite SAR data over the area of the Getty Museum, Los Angeles, as well as over an area around the city of Bam, Iran, stricken in 2003 by a serious earthquake. [ABSTRACT FROM AUTHOR]

A new growing method for simplex-based endmember extraction algorithms (EEAs), called simplex growing algorithm (SGA), is presented in this paper. It is a sequential algorithm to find a simplex with the maximum volume every time a new vertex is added. In order to terminate this algorithm a recently developed concept, virtual dimensionality (VD), is implemented as a stopping rule to determine the number of vertices required for the algorithm to generate. The SGA improves one commonly used EEA, the N-finder algorithm (N-FINDR) developed by Winter, by including a process of growing simplexes one vertex at a time until it reaches a desired number of vertices estimated by the VD, which results in a tremendous reduction of computational complexity. Additionally, it also judiciously selects an appropriate initial vector to avoid a dilemma caused by the use of random vectors as its initial condition in the N-FINDR where the N-FINDR generally produces different sets of final endmembers if different sets of randomly generated initial endmembers are used. In order to demonstrate the performance of the proposed SGA, the N-FINDR and two other EEAs, pixel purity index, and vertex component analysis are used for comparison. [ABSTRACT FROM AUTHOR]

In this paper, a new method to filter coherency matrices of polarimetric or interferometric data is presented. For each pixel, an adaptive neighborhood (AN) is determined by a region growing technique driven exclusively by the intensity image information. All the available intensity images of the polarimetric and interferometric terms are fused in the region growing process to ensure the validity of the stationarity assumption. Afterward, all the pixels within the obtained AN are used to yield the filtered values of the polarimetric and interferometric coherency matrices, which can be derived either by direct complex multilooking or from the locally linear minimum mean-squared error (LLMMSE) estimator. The entropy/alpha/anisotropy decomposition is then applied to the estimated polarimetric coherency matrices, and coherence optimization is performed on the estimated polarimetric and interferometric coherency matrices. Using this decomposition, unsupervised classification for land applications by an iterative algorithm based on a complex Wishart density function is also applied. The method has been tested on airborne high-resolution polarimetric interferometric synthetic aperture radar (POL-InSAR) images (Oberpfaffenhofen area—German Space Agency). For comparison purposes, the two estimation techniques (complex multilooking and LLMMSE) were tested using three different spatial supports: a fix-sized symmetric neighborhood (boxcar filter), directional nonsymmetric windows, and the proposed AN. Subjective and objective performance analysis, including coherence edge detection, receiver operating characteristics plots, and bias reduction tables, recommends the proposed algorithm as an effective POL-InSAR postprocessing technique. [ABSTRACT FROM AUTHOR]