It remains a challenging problem for data classification without training patterns. In many applications, there may exist some labeled data in other related domains (called source domain), and such labeled data can be helpful to solve the classification problem in the target domain. It is considered that the source domain and target domain are heterogeneous here and they represent the distinct feature spaces. A new transfer classification method for heterogeneous data is proposed based on the evidence theory. Some pattern pairs in the source domain and target domain are given to predict the link of these two domains. For each pattern in the target domain, we estimate its possible mapping value in the source domain using these pattern pairs with a self-organizing map (SOM) technique, and then the mapping value is classified using the labeled data in the source domain. However, the patterns with close values in the target domain may have more or less different values in the source domain due to the distinct characteristics of these two domains. So the mapping value can be very uncertain sometimes. In such a case, the target pattern is allowed to have multiple mapping values with different weights/reliabilities in the source domain. These mapping values can produce different classification results. The evidence theory is good at characterizing and combining uncertain information. In order to improve the classification accuracy, a new evidence-based weighted fusion method is developed for combining these classification results, which are discounted by the corresponding weights under the belief functions framework, and the final class decision is made according to the combination result. In experimental applications, some heterogeneous remote sensing data and UCI data are used to test the performance of new method with respect to several other methods, and it shows that the new method can efficiently improve the classification accuracy.

This paper aims to propose two main contributions in the field of multivariate data analysis through the possibility theory. The first proposition is the definition of a generalized family of multivariate elliptical possibility distributions. These distributions have been derived from a consistent probability-possibility transformation over the family of so-called elliptical probability distributions. The second contribution proposed by this paper is the definition of two divergence measures between possibilistic distributions. We prove that a symmetric version of the Kullback-Leibler divergence guarantees all divergence properties when related to the space of possibility distributions. We further derive analytical expressions of the latter divergence and of the Hellinger divergence for certain possibility distributions pertaining to the elliptical family proposed, especially the normal multivariate possibility divergence in two dimensions. Finally, this paper provides an illustration of the developed possibilistic tools in an application of bi-band change detection between optical satellite images.

A time series issued from modern synthetic aperture radar satellite imaging sensors is a huge dataset composed by many hundreds of million pixels when observing large-scale earth structures such as big forests or glaciers. A concise monitoring of these large scale structures for anomaly spotting thus requires loading and analyzing huge spatio/polarimetric multi-temporal image series. The contributions of the present paper for the sake of parsimonious analysis of such huge datasets are associated with a framework having two main processing stages. The first stage is the derivation of an index called geometric multi-wavelet total variation for fast and robust anomaly spotting. This index is useful for identifying significant abnormal patterns appearing as geo-spatial non-stationarities in multi-wavelet total variation map. The second stage consists in the proposal of a concise asymmetric multi-date change information matrix on regions associated with significant multi-wavelet total variations. This stage is necessary for a fine characterization of change impacts on existing geo-spatial structures. Experimental tests based on Sentinel-1 data show relevant results on a wide Amazonian forest surrounding the Franco-Brazilian Oyapock Bridge.

This paper presents a Credal algorithm to perform land cover classification from a pair of optical and radar remote sensing images. SAR (Synthetic Aperture Radar) /optical multispectral information fusion is investigated in this study for making the joint classification. The approach consists of two main steps: 1) relevant features extraction applied to each sensor in order to model the sources of information and 2) a Kohonen map-based estimation of Basic Belief Assignments (BBA) dedicated to heterogeneous data. This framework deals with co-registered images and is able to handle complete optical data as well as optical data affected by missing value due to the presence of clouds and shadows during observation. A pair of SPOT-5 and RADARSAT-2 real images is used in the evaluation, and the proposed experiment in a farming area shows very promising results in terms of classification accuracy and missing optical data reconstruction when some data are hidden by clouds.

Extracting information from a polarimetric radar representation usually consists in decomposing it with target decomposition algorithms. This first step can be seen as a geometric analysis of the polarimetric information: the identification of physical radar scattering mechanisms. The problem is that average physical parameters are estimated. As a consequence, these parameters might not describe correctly any of the land cover types that can be mixed together into the radar resolution cell. Therefore, using the polarimetric parameters for land cover classification is challenging. The novelty of the method is to propose a thematic analysis of the polarimetric information preceding the geometric one. The objective is to assess if splitting off polarimetric information on a land cover type basis before applying usual target decomposition algorithms can produce more consistent radar scattering mechanisms when land cover classes are mixed inside the radar resolution cell. A cooperative fusion framework in which very high-resolution optical images are used to unmix physical radar scattering mechanisms is proposed. For bare soil and forests, we point out that a linear unmixing model applied to the covariance matrix is able to split off polarimetric information on a land cover type basis. The assessment of the unmixed radar matrices is carried out with polarimetric radar images from the Radarsat-2 satellite. It was found that despite speckle, the reconstructed radar information after the unmixing process is statistically relevant with the observations. The question whether the unmixed radar images contain relevant thematic information is more challenging, but results tend to validate this property. This method could be used to have a better estimation of vegetation biomass in the context of open forested areas.

The change detection in heterogeneous remote sensing images remains an important and open problem for damage assessment. We propose a new change detection method for heterogeneous images (i.e., SAR and optical images) based on homogeneous pixel transformation (HPT). HPT transfers one image from its original feature space (e.g., gray space) to another space (e.g., spectral space) in pixel-level to make the pre-event and post-event images represented in a common space for the convenience of change detection. HPT consists of two operations, i.e., the forward transformation and the backward transformation. In forward transformation, for each pixel of pre-event image in the first feature space, we will estimate its mapping pixel in the second space corresponding to post-event image based on the known unchanged pixels. A multi-value estimation method with noise tolerance is introduced to determine the mapping pixel using -nearest neighbors technique. Once the mapping pixels of pre-event image are available, the difference values between the mapping image and the post-event image can be directly calculated. After that, we will similarly do the backward transformation to associate the post-event image with the first space, and one more difference value for each pixel will be obtained. Then, the two difference values are combined to improve the robustness of detection with respect to the noise and heterogeneousness (modality difference) of images. Fuzzy-c means clustering algorithm is employed to divide the integrated difference values into two clusters: changed pixels and unchanged pixels. This detection results may contain some noisy regions (i.e., small error detections), and we develop a spatial-neighbor-based noise filter to further reduce the false alarms and missing detections using belief functions theory. The experiments for change detection with real images (e.g., SPOT, ERS, and NDVI) during a flood in U.K. are given to validate the effectiveness of the proposed method.

In this paper we present a new similarity measure between possibility distributions based on the Kullback-Leibler (KL) divergence in the domain of real numbers. The possibility distributions are obtained thanks to the DFMP probability-possibility transformation [1] lying on the principle that a possibility measure can encode a family of probability measures. We consider here two particular possibility distributions built from parameter estimation of the Weibull and Rayleigh probability laws. The analytical expression of the KL divergence for the two considered possibility distributions are given, allowing a simple computation which depends on the parameters of the possibility distribution obtained. This new similarity measure is compared to the existing KL divergence for probability distributions in a context of change detection over simulated images as they provide a ground-truth of the changes required to evaluate the rate of true detection against false alarm.

This paper investigates the problem of glacier flow estimation using Synthetic Aperture Radar (SAR) image data. Our motivation is to exploit a weighted graph model constructed from characteristic points (i.e. keypoints) to measure the displacement vectors located at their positions. In fact, characteristic points are capable of capturing the image's radiometric and contextual information. Then, by encoding their interaction and inter-connection, a graph model is able to characterize both intensity and geometry information from the image content, which is relevant for texture tracking task. In this work, we employ a graph-based similarity measure to track the local texture information around each keypoint in order to figure out its correspondence from the other image and calculate the associated displacement. The proposed approach is tested and evaluated using high resolution TerraSAR-X images acquired from the Argentiere Glacier located in the French Alps. Our preliminary experimental results show the algorithm's capacity to provide a fast and reliable estimation of glacier flows, especially over highly textured and structured regions.

In classification problem, several different classes may be partially overlapped in their borders. The objects in the border are usually quite difficult to classify. A hybrid classification system (HCS) is proposed to adaptively utilize the proper classification method for each object according to the K-nearest neighbors (K-NNs), which are found in the weighting vector space obtained by self-organizing map (SOM) in each class. If the K-close weighting vectors (nodes) are all from the same class, it indicates that this object can be correctly classified with high confidence, and the simple hard classification will be adopted to directly classify this object into the corresponding class. If the object likely lies in the border of classes, it implies that this object could be difficult to classify, and the credal classification working with belief functions is recommended. The credal classification allows the object to belong to both singleton classes and sets of classes (meta-class) with different masses of belief, and it is able to well capture the potential imprecision of classification thanks to the meta-class and also reduce the errors. Fuzzy classification is selected for the object close to the border and hard to clearly classify, and it associates the object with different classes by different membership (probability) values. HCS generally takes full advantage of the three classification ways and produces good performance. Moreover, it requires quite low computational burden compared with other K-NNs-based methods due to the use of SOM. The effectiveness of HCS is demonstrated by several experiments with synthetic and real datasets.

Signal processing on graphs is a recent research domain that aims at generalizing classical tools in signal processing, in order to analyze signals evolving on complex domains. Such domains are represented by graphs, for which one can compute a particular matrix, called the normalized Laplacian. It was shown that the eigenvalues of this Laplacian correspond to the frequencies of the Fourier domain in classical signal processing. Therefore, the frequency domain is not the same for every support graph. A consequence of this is that there is no non-trivial generalization of Heisenberg's uncertainty principle, that states that a signal cannot be fully localized both in the time domain and in the frequency domain. A way to generalize this principle, introduced by Agaskar and Lu, consists in determining a curve that represents a lower bound on the compromise between precision in the graph domain and precision in the spectral domain. The aim of this paper is to propose a characterization of the signals achieving this curve, for a larger class of graphs than the one studied by Agaskar and Lu.

Deep Learning (DL), especially Convolutional Neural Networks (CNN), has become the state-of-the-art for a variety of pattern recognition issues. Technological developments have allowed the use of high-end General Purpose Graphic Processor Units (GPGPU) for accelerating numerical problem solving. They resort no only to lower computational time, but also allow considering much larger networks. Hence, nowadays computers are able to drive deeper, wider and more powerful models. State of the art CNNs have achieved human-like performance in several recognition tasks such as: handwritten character recognition, face recognition, scene labelling, object detection and image classification among others. Meanwhile, mobile devices have become powerful enough to handle the computations required for deploying CNNs models in near real-time. Here, we investigate the implementation of light-weight CNN schemes on mobile devices for domain-specific objection recognition tasks.

This paper proposes a novel approach for texture-based image indexing and retrieval in the scope of very high resolution (VHR) optical imagery. Our motivation is to take into account local textural features and structures inside each image to measure its similarity to other images. These local features are extracted for a set of characteristic points from the image using the local extrema-based descriptors (LED) from which the radiometric, spatial and gradient features of the local extrema pixels (i.e. maximums and minimums) are integrated to characterize local textures. Due to the fact that VHR images usually involve a variety of local textures which may weakly verify the stationarity hypothesis, an approach based on characteristic points like extrema pixels becomes relevant and effective. We perform our experimentation using texture databases extracted from VHR Pleiades images within the application of vineyard cultivation and oyster farming study. Retrieval results yielded by the proposed strategy are very promising and competitive compared to reference methods.

Wetlands, very sensitive and valuable ecosystem can be monitored in terms of water surfaces dynamics as well as vegetation characterisation and monitoring exploiting satellite data. The synergy between the recently launched Sentinel1 and Sentinel2 satellites have been investigated over the Poyang and Anhui lakes in PR China. Results highlight the gain in terms of operationality with a very high revisit exploiting the two systems, as well as for a thematic point of view with no was not yet reach at this resolution for water bodies and terrestrial, floating and submerged vegetation mapping and monitoring.

This paper studies the effects of stationary-based processing of moving ship signatures in synthetic aperture radar (SAR) imagery and introduces a methodology to estimate and compensate for them. SAR imaging of moving targets usually results in residual chirps in the azimuthal SLC processed signal. The fractional Fourier transform (FrFT) makes it possible to represent the SAR signal in a rotated joint time-frequency plane and performs optimal processing and analysis of these residual chirp signals. The along-track defocus can thus be compensated for and the target's azimuthal speed estimated. The impact of higher order motion terms (e.g., acceleration) has been also considered. Experiments were conducted on a large number of ship signatures extracted from Radarsat-2 Multi Look Fine and Ultra Fine SAR images. An intercomparison with a standard Doppler Sublook Decomposition Method (SDM) is carried out, as well as a complete performance analysis with AIS data as ground truth.

This paper investigates the problem of change detection in multitemporal synthetic aperture radar (SAR) images. Our motivation is to avoid using a large-size dense neighborhood around each pixel to measure its change level, which is usually considered by classical methods in order to perform their accurate detectors. Therefore, we propose to develop a pointwise approach to detect land-cover changes between two SAR images employing the principle of signal processing on graphs. First, a set of characteristic points is extracted from one of the two images to capture the image's significant contextual information. A weighted graph is then constructed to encode the interaction among these keypoints and hence capture the local geometric structure of this first image. With regard to this graph, the coherence of the information carried by the two images is considered for measuring changes between them. In other words, the change level will depend on how much the second image still conforms to the graph structure constructed from the first image. Additionally, due to the presence of speckle noise in SAR imaging, the log-ratio operator will be exploited to perform the image comparison measure. Experimental results performed on real SAR images show the effectiveness of the proposed algorithm, in terms of detection performance and computational complexity, compared to classical methods.

In this paper, a novel algorithm for textural feature description in very high resolution (VHR) satellite imagery is developed. It is based on a pointwise (PW) approach on the feature covariance matrix. The main motivation of this work is to construct the covariance matrix of oriented gradients (COG) using a nondense approach based on characteristic points (i.e., keypoints) extracted from the image. The proposed descriptor, which is named PW-COG, is expected to be effective when applied to VHR images. First, a COG-based descriptor is capable of not only capturing both radiometric and local geometric information (given by gradient features) from the image but also encoding their joint distribution and correlation, which are effectively relevant for texture characterization and discrimination. Second, by employing a keypoint-based approach, the proposed method is able to deal with large amount of VHR image data, since we do not take into consideration all pixels of the image, without requiring the stationarity hypothesis. In order to demonstrate the effectiveness of the proposed descriptor, texture-based image classification is carried out. Experimental study on both Brodatz texture database and VHR satellite images using the proposed algorithm provides very competitive results, in terms of texture discrimination and algorithm complexity, compared to reference methods.

In the framework of the evidence theory, several approaches for estimating belief functions are proposed. However, they generally suffer from the problem of masses attribution in the case of compound hypotheses that lose much conceptual contribution of the theory. In this paper, an original method for estimating mass functions using Kohonen's map derived from the initial feature space and an initial classifier is proposed. Our approach allows a smart mass belief assignment, not only for simple hypotheses but also for disjunctions and conjunctions of hypotheses. Thus, it can model at the same time ignorance, imprecision, and paradox. The proposed method for a basic belief assignment (BBA) is of interest for solving estimation mass functions problems where a large quantity of multivariate data is available. Indeed, the use of Kohonen's map simplifies the process of assigning mass functions. The proposed method has been compared with the state-of-the-art BBA technique on benchmark database and applied on remote sensing data for image classification purpose. Experimentation shows that our approach gives similar or better results than other methods presented in the literature so far, with an ability to handle a large amount of data.

From the multispectral top-of-atmosphere observations, ocean colour inversion aims at separating atmosphere and water contribution. In this context, we propose a novel Bayesian model with a focus on the definition of non-homogeneous priors on the aerosol and water multispectral signatures. The considered priors are set conditionally to observed covariates, typically geometry acquisition conditions and pre-estimates by a standard algorithm. We demonstrate from numerical experiments performed for real data the relevance of our non-homogeneous Bayesian setting to retrieve geophysically-consistent ocean colour images, in particular when dealing with complex coastal waters where standard algorithms perform poorly. Using a groundtruthed dataset, quantitative comparisons with operational schemes stress the overall improvement on the relative absolute error (respectively, 67% compared with the standard ESA MEGS algorithm and 9% compared with the ESA C2R neural network, for 12 bands ranging from 412 to 865 nm).

This study evaluates the potential of High Resolution Spotlight TerraSAR-X image for forest type discrimination. Emphasis is put on textural analysis accessible with high resolution radar data. Textural attributes are extracted from GLCM matrices, wavelet, and Fourier Transform (i.e. FOTO method). Their contribution for classification is assessed by their performance through the SVM algorithm.

A new method to unmix radar polarimetric images with optical images was proposed. This method has pointed out that the unmixing model is able to split off polarimetric information on a land cover type basis. In this paper unmixed radar polarimetric images obtained are compared with the observed ones in non-mixed conditions. Then, Cloude and Pottier decomposition is performed on the unmixed and observed radar images to asses whether the understanding of physical scattering mechanisms is improved with the unmixing. Finally, a classification experiment is designed to determine whether this fusion framework make the transfer of information from the optical images to the unmixed radar images possible.

Fusion of optical and radar remote sensing data is an important and challenging task for many applications such as multisource classification, due to the highly heterogeneous nature of the information they contain. The aim of this paper is to propose a novel strategy for multisensor optical-radar data classification in the evidential framework, which is well known for its ability to deal with imperfect information of each sensor and for its ability to exploit the redundancies and the complementarities between fused data and between sensors. A major contribution of our work is the use of Kohonen's map to model the input evidences of radar sensor since the optical sensor was addressed in our previous works. Also a specific design is presented here for fusing those heterogeneous data. First experiments show the effectiveness of the proposed approach.

This study addresses the performances of ship detection with data acquired by the newly launched Sentinel-1 SAR sensor. An automatic validation approach based on coastal AIS data is employed for measuring the detection efficiency. Results are compared with ship detection capabilities conducted on Radarsat-2 and CosmoSkymed datasets. The influence of different key parameters, such as SAR imaging characteristics (polarization, incidence angle) or meteorological conditions, is addressed. Such an analysis is useful for operational services to determine data specifications that assure optimum vessel detection for maritime surveillance applications.

This paper considers the impact of dynamical targets on SAR imagery, when processed with stationary based techniques. We propose to employ the Fractional Fourier Transform as a tool for estimating the residual Doppler rate corresponding to moving vessels. Hence, the defocusing effect can be corrected and the associated azimuthal velocity can be estimated. The capabilities of the proposed methodology are illustrated with moving vessels extracted from Radarsat-2 Multilook Fine images and AIS data flows as ground truth.

The present study involves an investigation of a pointwise approach on the feature covariance matrix to extract textu-ral features for very high resolution (VHR) satellite images. Indeed, our proposition is to construct the covariance matrix of oriented gradients using a non-dense approach based on characteristic points extracted from the image. This novel non-dense covariance descriptor is capable of not only capturing both radiometric and local geometric information from the image, but also encoding their joint distribution and correlation, which are effectively relevant for texture characterization and discrimination. In order to demonstrate the efficiency of the proposed descriptor, a texture-based image segmentation stage is carried out. First efforts on VHR panchromatic images using the proposed algorithm provide very promising and competitive results compared to classical methods.

The present paper proposes a texture-based unsupervised classification algorithm for fully polarimetric SAR (PolSAR) images. Here, the main motivation is to combine polarimetric information and local structure gradients from PolSAR image data to describe textural features and then use them for classification purpose. In this work, the notion of PolSAR image textures is characterized by two key features. First, the polarimetric coherency matrix is estimated using a weighted averaging operator based on patch similarity. Second, the image local geometry is taken into account by exploiting the structure gradient tensors. These characteristics are then integrated into texture descriptors via the approach of covariance matrix. Unsupervised classification stage is finally achieved by employing an adapted distance measure for covariance-based descriptors. Experiments performed on very high resolution complex PolSAR images using the proposed algorithm provide very promising results in terms of terrain classification and discrimination.

The classification of incomplete patterns is a very challenging task because the object (incomplete pattern) with different possible estimations of missing values may yield distinct classification results. The uncertainty (ambiguity) of classification is mainly caused by the lack of information of the missing data. A new prototype-based credal classification (PCC) method is proposed to deal with incomplete patterns thanks to the belief function framework used classically in evidential reasoning approach. The class prototypes obtained by training samples are respectively used to estimate the missing values. Typically, in a c -class problem, one has to deal with c prototypes, which yield c estimations of the missing values. The different edited patterns based on each possible estimation are then classified by a standard classifier and we can get at most c distinct classification results for an incomplete pattern. Because all these distinct classification results are potentially admissible, we propose to combine them all together to obtain the final classification of the incomplete pattern. A new credal combination method is introduced for solving the classification problem, and it is able to characterize the inherent uncertainty due to the possible conflicting results delivered by different estimations of the missing values. The incomplete patterns that are very difficult to classify in a specific class will be reasonably and automatically committed to some proper meta-classes by PCC method in order to reduce errors. The effectiveness of PCC method has been tested through four experiments with artificial and real data sets.

This study aims to evaluate the contribution of textural analysis from high spatial resolution TerraSAR-X images for tropical forests mapping. This study evaluates the potential of the High resolution Spotlight TerraSAR-X image (HS), with a spatial resolution of 0.6 meter in Range and 1.1 meter in Azimuth, for the classification of tropical forest plantation of native species. Indeed, the contribution of the analysis of textural information for classification has been emphasized through the analysis of Haralick textural parameters, a second order statistic parameters computed in a certain direction with a distance (d) and window size (w). The retained algorithm of classification is SVM (Support Vector Machine as it allows taking into account numerous parameters, which can be heterogeneous with respect to their physical dimension. To resolve the issue of class heterogeneity in the context of high resolution image, a post classification has been applied by the mean of a majority filter. In this case, the majority filter was weighted by integrating a shape of different plots containing the label of the tree planted species.

This paper caters to the land use and land cover monitoring with multitemporal and multipolarization radarsat data over an intensive agricultural area in France. The basic three objectives of this article are to evaluate the use of multitemporal sar data for land use and land cover classification in an area characterized by high spatiotemporal dynamics, evaluate the contribution of polarimetric information vs SAR data mono-polarisation and to evaluate “key data acquisition” for land use classification optimization in the study area.

A new method for unmixing radar polarimetric images with optical images is proposed. It was found that the polarimetric covariance matrix can be unmixed considering a linear model. As a result, this model is used to produce unmixed covariance matrices based on land cover types. We hope to prove that this unmixing of the polarimetric information produce greater information for land cover classification.

In the framework of the evidence theory, several approaches for estimating belief functions have been proposed. However, they generally suffer from the problem of masses attribution in case of compound hypotheses that lose much conceptual contribution of the theory. In this paper, an original method for estimating mass functions using the Kohonen map derived from the initial feature space and an initial classifier is proposed. Our approach allows a smart mass belief assignment, not only for simple hypotheses, but also for disjunctions and conjunctions of hypotheses. Thus, it can model at the same time ignorance, imprecision and paradox. The proposed method for basic belief assignment (BBA) is of interested for solving estimation mass functions problems where a large quantities of multi-variate data is available, such as remote sensing images. Indeed, the use of the Kohonen map simplify the process of assigning mass functions. In order to experimentally validate our work, we applied the proposed method to real data sets on image classification problem. Experimentation on SPOT images show that our approach gives better results than other methods presented in the literature.

This paper proposes a texture-based segmentation method for very high spatial resolution imagery. Indeed, our main objective is to perform a sparse image representation modeled by a graph and then to exploit the wavelet transform on graph for the final purpose of image segmentation. Here, a set of pixels of interest, called representative pixels, is first extracted from the image and considered as vertices for constructing a weighted graph. Once the wavelet transform on graph is generated, their coefficients serve as textural features and will be exploited for unsupervised segmentation. Experimental results show the effectiveness of the proposed method when applied for very high spatial resolution multi-spectral images in terms of good segmentation precision as well as low complexity requirement.

In this paper paper we address performances of ship detection algorithms in medium resolution SAR imagery. An automatic validation approach based on coastal AIS data allows to evaluate detectors efficiency. Detection capabilities remain sensitive to dataset features such as SAR imaging characteristics, meteorological conditions or vessel size. The influence of this different key parameters is fully assessed in this study. This analysis is valuable for operational services, allowing to select the most appropriate type of data for different applications in maritime surveillance.

We present a multidimensional evidential reasoning (MDER) approach to estimate change detection from the fusion of heterogeneous remote sensing images. MDER is based on a multidimensional (M-D) frame of discernment composed by the Cartesian product of the separate frames of discernment used for the classification of each image. Every element of the M-D frame is a basic joint state that allows to describe precisely the possible change occurrences between the heterogeneous images. Two kinds of rules of combination are proposed for working either with the free model, or with a constrained model depending on the integrity constraints one wants to take into account in the scenario under study. We show the potential interest of the MDER approach for detecting changes due to a flood in the Gloucester area in the U.K. from two real ERS and SPOT images.

In order to reconstruct missing data in very high resolution (VHR) multispectral images, several methodologies were proposed in the literature. However, missing data reconstruction still represents a complex image processing challenge to solve. A recent possibility comes from the compressive sensing (CS) theory, in particular the basis pursuit (BP) concept, which allows to find sparse signal representations in underdetermined linear equation systems. In this work, we propose an alternative selection method for the reconstruction of images adopting a histogram matching (HM) strategy. Experiments were conducted on FORMOSAT-2 images. The reported results include a simulation study and a comparison with a state-of-the-art technique for cloud removal.

Reconstructing missing data in very high resolution (VHR) multispectral images represents a complex image processing challenge. In this paper, we propose a new method for the reconstruction of areas obscured by clouds. It is based on compressive sensing (CS) theory, which allows to find sparse signal representations in underdetermined linear equation systems. Here we propose a novel implementation which exploits genetic algorithms (GAs) and a new strategy for the selection of atoms belonging to the dictionary. To illustrate the performances of the proposed method, a thorough experimental analysis on FORMOSAT-2 images is reported and discussed. It includes a simulation study and a comparison with a state-of-the-art technique for cloud removal.

A credal classification rule (CCR) is proposed to deal with the uncertain data under the belief functions framework. CCR allows the objects to belong to not only the specific classes, but also any set of classes (i.e. meta-class) with different masses of belief. In CCR, each specific class is characterized by a class center. Specific class consists of the objects that are very close to the center of this class. A meta-class is used to capture imprecision of the class of the object that is simultaneously close to several centers of specific classes and hard to be correctly committed to a particular class. The belief assignment of the object to a meta-class depends both on the distances to the centers of the specific class included in the meta-class, and on the distance to the meta-class center. Some objects too far from the others will be considered as outliers (noise). CCR provides the robust classification results since it reduces the risk of misclassification errors by increasing the non-specificity. The effectiveness of CCR is illustrated by several experiments using artificial and real data sets.

This letter presents an analysis of two synthetic aperture radar (SAR) images time series over a Sahelian area, which is located in the Gourma region, Mali. The first one is acquired at the C-band by the Environment Satellite-Advanced SAR sensor in wide swath (WS) mode and the second one at the L-band by the ALOS-Phased Array-type L-band SAR in fine, dual, and WS modes. A change detection method based on random projection and the hue-saturation-value transform appears to be very well appropriated to detect the temporal changes appearing on both data sets. The results illustrate the higher penetration depth of the L-band over such sandy soils. The C-band time series is found to be very sensitive to surface changes (soil moisture and vegetation growth). By contrast, the L-band time series appear to be very stable over the sandy soils, i.e., the only temporal changes occurring over temporary or permanent ponds.

Over the past few years, the reconstruction of missing data due to the presence of clouds received an important attention. Applying region-based inpainting strategies or conventional regression methods, such as support vector (SV) machine regression, may not be the optimal way. In this letter, we propose new combinations of kernel functions with which we obtain a better reconstruction. In particular, in the regression, we add to the radiometric information, i.e., the position information of the pixels in the image. For each kind of information adopted in the regression, a specific kernel is selected and adapted. Adopting this new kernel combination in a SV regression (SVR) comes out that only few SVs are needed to reconstruct a missing area. This means that we also perform a compression in the number of values needed for a good reconstruction. We illustrate the proposed approaches through some simulations on FORMOSAT-2 multitemporal images.

The paper provides a spatio-temporal change detection framework for the analysis of image time series. In this framework, the detection of changes in time is addressed at the image level by using a matrix of cross-dissimilarities computed upon wavelet and curvelet image features. This makes possible identifying the acquisitions of interest: the acquisitions that exhibit singular behavior with respect to their neighborhood in the time series, and those that are representatives of some stationary behavior. These acquisitions of interest are compared at the pixel level to detect spatial changes characterizing the evolution of the time series. Experiments carried out over European Remote Sensing (ERS) and TerraSAR-X time series highlight the relevancy of the approach for analyzing synthetic aperture radar image time series.

The intent of this paper is to propose new methods for the reconstruction of areas obscured by clouds. They are based on compressive sensing (CS) theory, which allows finding sparse signal representations in underdetermined linear equation systems. In particular, two common CS solutions are adopted for our reconstruction problem: the basis pursuit and the orthogonal matching pursuit methods. A novel alternative CS solution is also proposed through a formulation within a multiobjective genetic optimization scheme. To illustrate the performances of the proposed methods, a thorough experimental analysis on FORMOsa SATellite-2 and Satellite Pour l'Observation de la Terre-5 multispectral images is reported and discussed. It includes a detailed simulation study that aims at assessing the accuracy of the methods in different qualitative and quantitative cloud-contamination conditions. Compared with state-of-the-art techniques for cloud removal, the proposed methods show a clear superiority, which makes them a promising tool in cleaning images in the presence of clouds.

Very high resolution (VHR) images are appreciated for their high-level details, which significantly increase their application potential. However, typically, VHR images are affected by the presence of shadows. An attempt solution to the problem of shadows is to restore shadow-contaminated regions by compensating the value of shaded pixels. Unfortunately, it may happen that not all shadow areas are possible to restore. In this paper, we propose different criteria useful to help in understanding a priori if it is possible or not to reconstruct a specific shadow area. An ideal reconstructability criterion should not tolerate that an unreconstructable shadow area is assigned as reconstructable and, at the same time, should maximize the probability of detection of reconstructable areas. Several evaluation criteria working at the pixel and textural levels are presented. Furthermore, in order to select the best criteria, a fuzzy logic combination of the criteria is explored. A thorough experimental analysis is reported and discussed. It leads to the definition of a final global index based on the fusion of two single criteria, which are the Kullback–Leibler divergence and the angular second-moment difference.

The paper addresses multi-channel complex image filtering. It provides regularization cost functions associated to non-conventional vector and matrix iv norms for promoting geometry properties. The approach is shown to be efficient for filtering PolSAR images.

In this paper, a new method for three-dimensional building extraction from building specific low-level features is presented. This method, which addresses both building detection and building parameter estimation, relies on a maximum likelihood model inversion performed from low-level primitives, after they have been extracted from a single SAR HR amplitude image. The effectiveness of the proposed approach, which has been designed so as to robustly deal with noisy primitives, is demonstrated on a TerraSar-X image.

Reconstructing missing data in very high resolution (VHR) multispectral images represents a complex image processing challenge. In this paper, we propose a new method for the reconstruction of areas obscured by clouds. It is based on compressive sensing (CS) theory, which allows to find sparse signal representations in underdetermined linear equation systems. In particular a common CS solution is adopted for our reconstruction problem: the orthogonal matching pursuit (OMP) method. To illustrate the performances of the proposed method, a through experimental analysis on FORMOSAT-2 multispectral images is reported and discussed. It includes a simulation study and a comparison with a state-of-the-art technique for cloud removal.

A typical issue encountered in very high resolution (VHR) optical imagery is the undesired presence of shadows. Their compensation by means of appropriate techniques is however not always successful. In this work, we present different criteria, which aim at understanding a priori if it is possible or not to reconstruct a specific shadow area. They are based on a quantitative comparison between the histograms characterizing the shadow and non-shadow areas of the same typology of ground cover. A thorough experimental analysis performed on three different images is reported and discussed.

In the framework of the monitoring of structures and infrastructures from environmental disasters, the COSMO-SkyMed constellation has a huge potential, thanks to up to metric spatial resolution, short revisit time, and the day/night all-weather acquisition capability ensured by SAR. This paper focuses on the scientific results of the project “Development and validation of multitemporal image analysis methodologies for multirisk monitoring of critical structures and infrastructures,” funded by the Italian Space Agency. Several change-detection, data-fusion, and feature-extraction techniques, which were developed and experimentally validated in the project for COSMO-SkyMed imagery and for their integration with other data sources (including very high resolution optical data), are described and examples of processing results are discussed.

So far, sea surface scattering polarization sensitivity is still difficult to fully understand. To gain understanding, full polarization C-band data spectral analysis is considered and the de-polarization (HV) and polarization ratio (PR) are jointly analysed with in-situ wind and wave observations.

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.

The presence of shadows in very high resolution (VHR) images can represent a serious obstacle for their full exploitation. This paper proposes to face this problem as a whole through the proposal of a complete processing chain, which relies on various advanced image processing and pattern recognition tools. The first key point of the chain is that shadow areas are not only detected but also classified to allow their customized compensation. The detection and classification tasks are implemented by means of the state-of-the-art support vector machine approach. A quality check mechanism is integrated in order to reduce subsequent misreconstruction problems. The reconstruction is based on a linear regression method to compensate shadow regions by adjusting the intensities of the shaded pixels according to the statistical characteristics of the corresponding nonshadow regions. Moreover, borders are explicitly handled by making use of adaptive morphological filters and linear interpolation for the prevention of possible border artifacts in the reconstructed image. Experimental results obtained on three VHR images representing different shadow conditions are reported, discussed, and compared with two other reconstruction techniques.

A novel approach for building extraction in high resolution optical and SAR images and its application to building change detection are presented in this paper. The proposed object-based building extraction technique relies on an hypothesis generation-optimization-validation scheme, whose estimation and detection performances are characterized on Quickbird and TerraSAR-X semi-urban images. The introduced building extraction approach is then applied to building change detection in a pair of high resolution heterogeneous images.

Soil salinization is one of the most hazardous phenomenon accelerating the land degradation processes. Map ping and tracking soil salinity changes is fundamental for anticipating natural disaster, such as desertification, in arid and semi-arid regions. In this work, we establish an empirical model for soil salinity mapping based on a gaussian mixture and using field electrical conductivity (EC) measures. The developed model is tested on saline soil samples collected from the semi-arid region of Kairouan located in central Tunisia. It is based on statistical moments derived from multiband (HH and VV) intensity synthetic aperture radar (SAR) data of the Envisat satellite. The resulting salinity map is composed of three classes of salinity (Low, Medium and High) with respect to the EC measurements. The developed model is validated for low salinity distribution, whereas, it needs more samples to be generalized for medium and high soil salinity content.

Reconstructing missing data in very high resolution (VHR) multispectral images represents a complex image processing challenge. In this paper, we face this problem through a novel solution based on inpainting. Inpainting is a technique to reconstruct missing regions in a given image by propagating the spectro-geometrical information retrieved from the remaining parts of the image. Our proposed solution relies on the idea to formulate the inpainting process under a multiresolution processing scheme. Experiments conducted on a VHR image are reported and discussed.

Missing data in very high spatial resolution (VHR) optical imagery take origin mainly from the acquisition conditions. Their accurate reconstruction represents a great methodological challenge because of the complexity and the ill-posed nature of the problem. In this letter, we present three different solutions, with all based on the inpainting approach, which consists in reconstructing the missing regions in a given image by propagating the spectrogeometrical information retrieved from the remaining parts of the image. They rely on the idea to enrich the patch search process by including local image properties or by isometric transformations or to reformulate it under a multiresolution processing scheme, respectively. Thorough experiments conducted on two different VHR images are reported and discussed.

Within the framework of the unmixing of hyperspectral images, the pixel mixture is a difficult problem to solve. This difficulty comes from several outliers which seriously affect the reliability of spectral unmixing results. The illumination change effect, where the image does not reflect the true appearance of the scene, in many cases due to shadow facts, is considered to be one of the most important outliers. The present work proposes a new approach called Spectral Angle Measure-based Spectral Unmixing which uses the spectral angle constraint for abundance quantification. The major benefit of this approach is its ability to take advantage of the geometric properties of the Spectral Angle Measure technique to estimate abundance quantification independently of the amplitude (magnitude) of the Endmembers spectral signatures, using only spectral angle measures. As a consequence, a significant reduction in spectral unmixed error corresponding to the spectral similarity within-class confusion is obtained. A second benefit concerns physical constraints which are respected. The experiment was conducted using simulated and real images to validate our approach and to compare it with a well known statistical one.

The aim of this paper is to present a general framework for change detection in a time series of radar images, for an operational purpose and in the context of environmental monitoring. The change detection procedure is turned into the framework of detecting a random signal into the noise; the detection of this signal leads to the detection of a change in the time series. This framework is based on a non-parametric detection method that assume a sparse representation of the data. When using radar images, the speckle noise invalidates the hypothesis of sparsity. Then a pre-processing technique is required to provide an appropriate sparse representation of data, whatever the initial noise characteristics. The paper focuses on the change indicator, based on recursive median filtering, yielding a piecewise regular representation of a scene obtained by spreading the statistically most reliable pixel values over the image. The recursive median filtering leads to simple change indicators that are more efficient than the Kullback-Leibler change indicator when using small analyzing sliding window. Furthermore, it induces an simple extension to perform progressive change characterization through a multi-temporal filtering approach. Results are shown with a two-date change detection from RADARSAT images and from a time series of ERS and ENVISAT images.

This paper provides asymptotic properties of the autocorrelation functions of the wavelet packet coefficients of a fractional Brownian motion. It also discusses the convergence speed to the limit autocorrelation function, when the input random process is either a fractional Brownian motion or a wide-sense stationary second-order random process. The analysis concerns some families of wavelet paraunitary filters that converge almost everywhere to the Shannon paraunitary filters. From this analysis, we derive wavelet packet based spectrum estimation for fractional Brownian motions and wide-sense stationary random processes. Experimental tests show good results for estimating the spectrum of1/fprocesses.

The paper presents a general framework for change detection in radar images, for an operational purpose and in the context of environmental monitoring. This framework is based on a processing which provides highly sparsifiable representations of data. This processing is called turbo-median and is a combination of the sample median robustness and the turbo principle for iteratively correcting errors. The turbo-median processing of a scene is an homogenized representation based on an iterative median and which consists in spreading the statistically more robust measurements of the scene under consideration over the size of the image representing this scene. It allows for reducing the change detection problem into the problem of detecting a signal, with unknown distribution, in additive noise.

A method for modeling and tracking convective clouds within radar images is presented. An object modeling approach is used, based on the extraction of grayscale skeletons from 2-dimensionnal cross-section of 3-dimensional radar data. Grayscale skeletons are appropriate shape descriptors for non-rigid and heterogeneous objects, in which gray-level local maxima correspond to regions of interest. The modeling scheme is enhanced by linking meta-data to some chosen points of the skeleton; this provides a good representation of the weather scene in terms of hazards for an aircraft. Skeletons are stored within a graph structure and tracked among successive pictures by means of relaxation labeling processes. The deduced ad-vection field is used to nowcast the clouds evolution. Preliminary results are satisfying concerning advection forecast.

This paper focuses on the flexibility of a multidimensional model of probability density function (pdf) to describe distribution of complex data in polarimetric SAR images. This model is based on Copulas Theory for characterizing the dependence between the polarimetric channels (HH, VV, HV, VH). This corresponds to finding a model based on multidimensional copulas to describe the behavior of the target vector. The advantage in using copulas theory is to extend correlation concept to a wider dependence one, which may be non-linear, especially when processing high-resolution data. So, from this point of view, the model is more flexible than the classical Wishart distribution since no speckle filtering is required as preprocessing step to model accurately the pdfs. The other advantage of copulas is to split dependence concept and marginal distributions. Then, this multidimensional characterization may be linked to pdf which are not necessary of circular Gaussian law. So, specific parametric distribution may be choosen to fit each component (modulus and phase) of the Sinclair matrix. It yields a flexible model, for characterizing statistical behavior of the polarimetric SAR data, that may be derived to produce a segmentation algorithm.

This paper focuses on the implementation of the Kohonen's Self Organizing Map (SOM) through the Orfeo Toolbox. It makes the link between the theoretical background and the strategic choices in the implementation: (1) Generic choice of the learning functions that ensures convergence of the training process; (2) Generic definition of the neuron that allows flexible use on multicomponent data, time series and also heterogeneous data; (3) Flexible class inheritance to redefine the update function to make the SOM deal with missing data. Example are been given with the preprocessing of low resolution time series contaminated by clouds and shadows due to weather conditions during the acquisitions. The technique is capable to reconstruct a complete time series free of clouds from MODIS data. It will also focus on the interest of a sparse distance criteria to reconstruct the time series on presence of haze that is difficult to detect.

This paper presents a new approach for unsupervised change detection in pairs of Synthetic Aperture Radar (SAR) images. As changes to detect can have various sizes and intensities which are a priori unknown in most applications, we propose a multiscale approach without considering any a priori information. Using multiscale series of a cumulant-based Kullback-Leibler divergence (CKLD) measure computed between two dates, changes are characterized as areas where the CKLD values vary a lot when the scale varies. In a probabilistic a-contrario framework, a measure of meaningfulness of such an evolution through scale is derived, leading to a criterion free of parameter. Results are presented using a pair of SAR images acquired before and after the volcanic eruption of the Nyiragongo in January 2002 (Congo), showing the robustness of the method with respect to the number of false alarms.

Satellite imagery classification using the support vector machine (SVM) algorithm may be a time-consuming task. This may lead to unacceptable performances for risk management applications that are very time constrained. Hence, methods for accelerating the SVM classification are mandatory. From the SVM decision function, it can be noted that the classification time is proportional to the number of support vectors (SVs) in the nonlinear case. In this letter, four different algorithms for reducing the number of SVs are proposed. The algorithms have been tested in the frame of a change detection application, which corresponds to a change-versus-no-change classification problem, based on a set of generic change criteria extracted from different combinations of remote sensing imagery.

This paper presents a new sigmoid-based wave shrink function. The shrinkage obtained via this function is particularly suitable to reduce noise without impacting significantly the statistical properties of the signal to be recovered. The proposed WaveShrink function depends on a parameter that makes it possible to control the attenuation degree imposed to the data, and thus, allows for a flexible shrinkage.

Due to their flexibility, and capacity to handle high dimensional vectorial data, support vector machines (SVMs) have become the reference for remote sensing imagery classification. However when processing large amounts of data the SVM classification could be a time consuming process. In this paper a new decomposition scheme of the SVM decision function is proposed. The decomposition is based on using the Taylor series expansion to approximate the kernel function. Then, using the results of the optimization problem of the SVM after the learning phase, this expansion is used to obtain an approximate decision function that provides a trade-off between the classification accuracy and the processing time. This speeds-up the SVM classification if limited processing time is available and favors accuracy if sufficient processing time is available.

The problem of focusing on the most relevant information in a potentially overwhelming quantity of data has become increasingly important. Using irrelevant or noisy features not only can affect the accuracy of the classification results obtained but also the convergence time. In this paper several feature selection algorithms used with the Support Vector Machine (SVM) algorithm are presented. The feature selection algorithms are classified as filter and wrapper approaches. Two different wrapper techniques are presented: the first one uses the generalization error estimate of the leave-one-example-out error, while the second one uses the error estimate of the leave-one-feature-out error. Filter approaches with 4 different parameters are presented, namely: the mutual information, the FScore, and two advanced entropy measures are studied. Results in the context of change detection using satellite imagery are then discussed.

In the context of change detection and due to the multitude of change scenarios, the objective is to build a generic change detection system. For many technical and operational reasons the Support Vector Machines (SVM) algorithm is used. One of the crucial steps when using the SVM algorithm is the choice of the kernel function. With the lack of a priori information the choice of the kernel function may be difficult for the user. In this paper several techniques for constructing a suitable kernel function obtained from the data are proposed.

Spaceborne Synthetic Aperture Radar (SAR) is well adapted to detect ocean pollution independently from daily or weather condition. As it is sensitive to surface roughness, the presence of oil film on the sea surface decreases the backscattering of the sea surface resulting in a dark feature patches in SAR images. In fact, oil slicks have specific impact on ocean wave spectra. Initial wave spectra may be characterized by three kinds of waves, big, medium and small, which correspond physically to gravity and gravity-capillary waves. The increase of viscosity due to the presence of oil damps gravity-capillary waves. This induces a damping of the backscattering to the sensor, but also a damping of the energy of the wave spectra, then it modifies the sea surface roughness observed by the sensor. Thus, local detection of wave spectra modification may be achieved by a appropriated texture analysis of the original SAR image. In this paper, the texture analysis is based on measure of similarity between a local probability density function (pdf) of clean water and the local pdf of the zone to be inspected. The local distribution is estimated in the neighbourhood of each pixel, through a sliding window, and compared to the reference one by using the Kullback-Leibler (KL) distance between distributions. An efficient strategy has been adopted in order to perform pdf estimation through a non-parametric approach.

Bi-date change detection and image registration are based on local similarity measures. When applied to Synthetic Aperture Radar (SAR) observations, the first one is based on the comparison of the local probability density functions (pdf) of the 2 images, while the latter is based on correlation or mutual information (MI) measure. A specific implementation of MI has been found to be efficient in change detection from heterogeneous SAR images as well as for SAR image registration. This implementation may be splitted in 2 terms and that can be linked to the change detection part and to the registration one respectively. Hence a set of measures are proposed in order to perform similarity measure for change detection in homogeneous and heterogeneous SAR images that prevents from misregistration errors. This point of view has been applied to a set of Radarsat images and a pair or ERS images acquired in the frame of the International Charter Space and Major Disasters, corresponding to a lava flow and a flooding event.

Monitoring changes in the vegetation cover during the intercrop season is of special interest in intensive agricultural region, such as the Brittany region in France, to locate bare soils and control their influence to the environment. The presence of bare soils leads to detrimental environmental effects such as soil erosion or water quality degradation. Therefore, identification and monitoring of bare soils at a regional scale in the winter season are required for any agricultural management program. Data from the Moderate Resolution Imaging Spectroradiometer have been selected for this paper due to their low spatial resolution, which decreases the cost of processing and storage, and the high revisit frequency, which increases the probability to acquire scenes free of clouds and shadows during the winter season. Unfortunately, few images per season only are free of cloud contamination and associated shadows. Therefore, the specific objective of this paper is to develop and implement a preprocessing method to recover spectral values of contaminated data by weather conditions for subsequent bare-soil mapping. In the context of this paper, Kohonen's self-organizing map (SOM) is used for recovering data contaminated by weather conditions that have been considered as erroneous data. This nonparametric regression procedure has proved its success to deal with missing-values problem. Hence, the erroneous-values problem, reflectance values contaminated by clouds or shadows, has been converted to the missing-values problem by using a cloud and shadow (outlier) detector. The SOM algorithm was tested also on the erroneous data directly, but better results were found with the “missing values” formulation. The idea is to, first, train SOM onto clear temporal profiles free of clouds and shadows during the winter season. Second, erroneous values are converted to missing values by an outlier detector which operates on each temporal profile (set of colocated pixels acquired at different dates). Finally, the SOM algorithm for missing values is used to estimate contaminated reflectance values.

A new preprocessing technique is presented in this paper to automatically highlight changes in multitemporal strongly heterogeneous remotely sensed images. The proposed technique is devoted to the case where the two acquisitions, before and after a given event, are significantly different, due, for instance, to different sensors, acquisition modalities, or climatic conditions. In a previous study, it was proven that the local statistics of the images acquired at the two dates could be used to extract a relevant change indicator. Nevertheless, this measure is valid when the two observations have been derived from similar acquisitions. When the acquisition modalities differ, local statistics tend to be too different from one image to the other one to be relevant in highlighting the ground evolution without mixing with the changes at ground. The technique proposed in this paper to overcome this limitation is based on the assumption that some dependence indeed exists between the two images in unchanged areas. This dependence is modeled by quantile regression applied according to the copula theory and used to perform an estimation of the local statistics that would have been observed if the acquisition conditions of the first image had been similar to the ones of the second image. The method yields an estimate of the local statistics of the first image through the point of view of the second one. Then, usual Kullback-Leibler-based comparisons of those statistics are applied to define a change measure, which may be analyzed (e.g., by thresholding) in order to detect changes. Experimental results are shown to validate the proposed method by using a pair of Synthetic Aperture Radar (SAR) images onboard European Remote Sensing (ERS) Satellite images and a pair of optical-SAR images (from the High Resolution Visible (HRV) sensor onboard Satellite Pour l'Observation de la Terre (SPOT) satellite and from ERS-SAR) acquired before and after a flood.

In the framework of the International Charter "Space and Major Disasters", charter calls are made to the signing parties every time a natural or technological hazard occurs. Consequently, space data are provided by the partners in order to help local authorities to assess the damages, organize and optimize the use of available resources. In such cases, abrupt change detection algorithms are required and numerous methods have been proposed by the geoscience and remote sensing (GRS) community. In this paper statistics measured on charter calls are compared to statistics measured on change detection methods published in the literature. These statistics aim to give the image processing community a better understanding of the needs and challenges faced in the case of real life disaster scenarios.

Monitoring changes in the vegetation cover during the intercrop season is of a special interest in intensive agricultural regions. The presence of bare soils leads to detrimental environmental effects such as soil erosion or water quality degradation. To identify and monitor winter land cover at a regional scale in the Brittany region in France, data from the Moderate Resolution Imaging Spectroradiometer (MODIS) low resolution time series were used. This study examines the Self-Organizing Map (SOM) for classifying winter land cover with MODIS time series in agricultural and fragmented landscapes. Field data and high resolution images have been used to generate two sets of pixels for training and validation. A comparison between the suggested method and NDVI thresholding is presented.

The paper focuses on a flexible model of multidimensional probability density function (pdf) dedicated to describe amplitude distribution of polarimetric SAR data. The model is based on the copula theory for characterizing the dependency between polarimetric channels (HH, VV, HV/VH or the target vector components). The benefit in using copula theory is to extend correlation concept to a wider dependence one, which may not be linear. From this point of view, the model is more flexible than the classical Wishart distribution. But it may include it. The other benefit in using the copula model is to separate the dependence concept from the shape of the marginal pdfs. Hence, this multidimensional characterization may be linked to classical 1D gamma pdf, or to a more flexible Pearson system of distributions. In the case of high resolution data, pdf shapes are becoming of heavy tailed and the Fisher system of distributions seems to be an interesting alternative for such a model. Any parametric 1D model may be used. The paper mainly focuses on the model itself and more precisely on the technique required to construct such multidimensional dependence function. The difficulties arise for copula on 3D in which the dependency is not homogeneous between the components (the link between HH and VV may not be of the same behavior as the one between HH and HV). Illustrations are given on classification and despeckling. Classification will be performed by a stochastic estimation maximisation (SEM). Despeckling will be achieved by a maximum A posteriori technique.

A new preprocessing technique is presented in this paper to perform automatic change detection in multitemporal multimodal remotely sensed images, mainly synthetic aperture radar (SAR) ones. This technique is dedicated to the case where the two acquisitions, before and after an major disaster, are different for some reason (different sensor, modality of acquisition or climatic conditions). A measure, based on the local statistics of the images between the two dates, has proved to be a relevant change indicator. Nevertheless, the measure is valid when the two observations have been acquired with a similar point of view only. When the modalities of acquisition differ, local statistics tend to be too different, from one image to the other, to be relevant to the ground evolution without mixing to the normal changes. The technique, that overcomes this constraint, is based on the assumption that some dependence exists indeed between the two images. This dependence is modelled by the copula theory and used to perform an estimation of the local statistics that would have been observed if the modality of the first image had been similar to the other. It yields an estimation of local statistics of the first image, through the point of view of the latter. Then, usual comparison of those statistics may be applied to perform change detection. Some results are shown on a pair of ERS images and pairs of SPOT/ERS acquired before and after a flood.

In this paper, we present a new similarity measure for automatic change detection in multitemporal synthetic aperture radar images. This measure is based on the evolution of the local statistics of the image between two dates. The local statistics are estimated by using a cumulant-based series expansion, which approximates probability density functions in the neighborhood of each pixel in the image. The degree of evolution of the local statistics is measured using the Kullback-Leibler divergence. An analytical expression for this detector is given, allowing a simple computation which depends on the four first statistical moments of the pixels inside the analysis window only. The proposed change indicator is compared to the classical mean ratio detector and also to other model-based approaches. Tests on the simulated and real data show that our detector outperforms all the others. The fast computation of the proposed detector allows a multiscale approach in the change detection for operational use. The so-called multiscale change profile (MCP) is introduced to yield change information on a wide range of scales and to better characterize the appropriate scale. Two simple yet useful examples of applications show that the MCP allows the design of change indicators, which provide better results than a monoscale analysis

In this paper, we present a new similarity measure for automatic change detection in multitemporal SAR images. This measure is based on the evolution of the local statistics of the image between two dates. The local statistics are estimated using a cumulant-based series expansion which approximates the probability density functions in the neighborhood of each image pixel. The degree of evolution of the local statistics is measured using the Kullback-Leibler divergence. An analytical expression for this detector is given allowing a simple computation which depends only on the 4 first statistical moments of the pixels inside the analysis window. The concept of multiscale change profile (MCP) is also introduced and its optimized implementation is presented. MCP yields change information on a wide range of scales and better characterizes the appropriate scale to be used for the detection. Two simple examples of application show that the MCP allows the design of change indicators which provide better results than a monoscale analysis.

In the context of this study, changes in the vegetation coverage are to be detected in the Brittany region in France, based on seasonal and regional scales. Thus, many low resolution images from MODIS satellite are considered per season. Unfortunately, the Brittany region in the west of France is often covered by clouds, and few dates per season with clear coverage are available. A nonparametric regression procedure, depending on the presence of clear small zones during the season, for predicting the value of data contaminated by weather conditions is presented. The problem is the erroneous data (pixels contaminated by clouds or shadows), which may be seen as a problem of an incomplete data as soon as an outlier detection technique is properly chosen and applied. The idea is to use the Kohonen's self organizing map (SOM) onto clear zones (free of clouds and shadow) for training. Then, incomplete data may be recovered by projecting the data onto this trained SOM. To detect pixels contaminated by clouds and shadows, the box and whisker method has been used with an overall accuracy of 97.7636% and a kappa coefficient of 0.9652.

This paper shows how to obtain a binary change map from similarity measures of the local statistics of images before and after a disaster. The decision process is achieved by the use of a zz-SVM in which a stochastic kernel has been defined. Stochastic kernel includes two similarity measures, based on the local statistics, to detect changes from the images: 1) A distance between maginal probability density functions (pdfs) and 2) the mutual information between the two observations. Distance between marginal pdfs is evaluated by using a series expansion of the Kullbak-Leibler distance. It is achieved by estimating cumulants up to order 4 from a sliding window of fixed size. Mutual information is estimated through a parametric model that is issued from the copulas theory. It is based on rank statistics and yields an analytic expression, that depends on the parameter of the copula only, to be evaluated to obtain the mutual information. Preliminary results are shown on a pair of Radarsat images acquire before and after a lava flow. A ground truth allows to show the accuracy of the stochastic kernels and the SVM decision.

Spaceborne synthetic aperture radar (SAR) is well adapted to detect ocean pollution independently from daily or weather conditions. In fact, oil slicks have a specific impact on ocean wave spectra. Initial wave spectra may be characterized by three kinds of waves, namely big, medium, and small, which correspond physically to gravity and gravity-capillary waves. The increase of viscosity, due to the presence of oil damps gravity-capillary waves. This induces not only a damping of the backscattering to the sensor but also a damping of the energy of the wave spectra. Thus, local segmentation of wave spectra may be achieved by the segmentation of a multiscale decomposition of the original SAR image. In this paper, a semisupervised oil-slick detection is proposed by using a kernel-based abnormal detection into the wavelet decomposition of a SAR image. It performs accurate detection with no consideration to signal stationarity nor to the presence of strong backscatters (such as a ship). The algorithm has been applied on ENVISAT Advanced SAR images. It yields accurate segmentation results even for small slicks, with a very limited number of false alarms

Efficient observation means are required for regional-scale detection of oil slicks at sea, as well as for local-scale quantitative mapping in order to support operational fight and recovering operations, including reliable choice and guidance of maritime and airborne fighting means. An efficient oil slick detection algorithm based on a multiscale approach is proposed for operational regional-scale detection from satellite SAR images. The potential of combining airborne passive hyperspectral imagery and active fluorescence laser technology is proposed for local-scale quantitative characterization. The ways towards the use of both satellites and airborne remote sensors for use in operational emergency scenarios are discussed.

Spaceborne synthetic aperture radar is well adapted to detect ocean pollution independently from daily or weather condition. In fact, oil slicks have specific impact on ocean wave spectra. Initial wave spectra may be characterized by three kinds of waves, big, medium and small, which correspond physically to gravity and gravity-capillary waves. The increase of viscosity due to the presence of oil damps gravity-capillary waves. This induces a damping of the backscattering to the sensor, but also a damping of the energy of the wave spectra. Thus, local segmentation of wave spectra may be achieved by the segmentation of a multiscale decomposition of the original SAR image. In this work, an unsupervised oil slick detection is proposed by using kernel-based novelty detection into the wavelet decomposition of a SAR image. It performs accurate detection with no consideration to signal stationarity nor to the presence of strong backscatters (such as ships). The algorithm has been applied on ENVISAT ASAR images. It yields accurate segmentation results, even for small slicks, with a very limited number of false alarms.

Covering soils with vegetation during the fallow and planting seasons is one of the main ways to reduce water pollution, by restricting pollutant fluxes to aquatic systems. The bare soil/vegetation ratio monitoring can be carried out daily with a coarse spatial resolution using SPOT VEGETATION (1 km). Nevertheless, land-cover changes detected at a regional scale with this ratio may be due to winter vegetation cover changes as well as the influence of climatic events. Therefore, observed changes have to be validated from a local-scale analysis with higher spatial resolution data. The aim of this study is to develop a technique that allows high or low variations detected at a regional scale to be assessed from SPOT VEGETATION images with data acquired at a higher scale, SPOT High Resolution Visible and Infrared images in our case. In this study, the link between the images from the two sensors is achieved from the design of an artificial neural network method based on a Kohonen self-organizing map. The originality of this method lies in the use of external knowledge from ground observations and the use of temporal behavior to solve such a change of scale. Results of testing this method by using a potential change map based on the last few years' land-cover observations have shown a good correspondence between the observed and predicted bare soil/vegetation balance with regards to the spatial resolution difference between the two sensors.

The present work develops a new aspect in the use of HF radar as a tool for real time operational oceanography. We propose to implement new algorithms for detecting anomalies (outliers) on VHF radar data. Perspectives of such work are to detect changes in sea surface, such as ship or oil pollution. In this preliminary work, no information are given about the anomalies. We propose to study new detection methods based on Support Vector expansion to perform a binary separation between observations. This technique is well adapted to HF radar observations that are never statistically representative but give a good temporal resolution

C-band SAR is well adapted to detect ocean pollution because backscatter is reduced by oil slicks. They appear as dark patches on the image as the increase of viscosity due to the presence of oil damps gravity-capillarity waves. In order to detect these dark patches, we use algorithms based on filters, gradients, and morphological mathematics, and a new approach based on ocean surface characterization. We have tested these methods on ERS and ENVISAT images acquired during Prestige tanker wreckage and the results are compared with aircraft surveys. We conclude that slicks with high contrast and simple shape are easily detected using basic algorithms, but most of the time, other methods are needed. The ocean characterization method is a way to follow for improving oil slick detection and providing decision aids for classification step

The problem of detecting abrupt changes in a set of synthetic aperture radar (SAR) images is carried out by comparing the results of segmentation between images with different modalities acquired before and after a disaster. Individual segmentations are not considered by themselves since they yield surface map characterization while the goal is to detect a temporal evolution of soil characteristics. Hence, we propose to use the modification of class statistics in the images in order to characterize potential changes and to prevent from false alarms that may be induced by the specific modality of each SAR acquisition. The change detection process is divided in two steps; 1) segmentation of the observations in order to have an estimation of the marginal probability distribution function (pdf) of each class; 2) comparison of the pdf from different images to detect changes by means of evidential and paradoxical reasoning. This two-stages process has been applied on Radarsat images (F2 and F5) of the Nyiragongo volcano, DR Congo, erupted on January 2002. The results obtained outperform simple strategies based on image differencing/ratioing

This work extents the hidden Markov chain (HMC) model for the unsupervised segmentation of multicomponent images. Although the vectorial extension of the model is almost straightforward, we are faced to the problem of estimating a mixture of nonGaussian multidimensional densities. In this work, we adopt an independent component analysis (ICA) approach that allows the mutual dependance between the layers to be taken into account in the segmentation process. Classification results on a four bands SPOT-IV image illustrates the method. Also, a comparison is performed when only mutual independence or correlation between the components is assumed.

Leaving fields with a vegetation cover during the winter is one of the main ways to reduce water pollution, in restricting pollutant fluxes towards rivers. The bare soils/vegetation ratio monitoring can be carried out daily at a coarse spatial resolution with SPOT VEGETATION (1 km), and also at a higher spatial resolution with SPOT HRVIR (20 m), but with less repetitive and spatially more restricted data. Land-cover changes detected at a regional scale with this ratio can be explained by winter vegetation covering changes as well as by influence of climatic events. Therefore, observed changes have to be validated from a local scale analysis. The aim of this study is to develop a method that allows to assess high or low variations detected at a regional scale from SPOT VEGETATION images with data registered at a higher scale, SPOT HRVIR images in our case. In this study, the link between the images of the two sensors is set up from the design of an Artificial Neural Network method based on a Kohonen Self-Organizing Features Map. The originality of this method lies in the use of temporal dimension to solve such a change of scales.

A new method is proposed in this paper to detect oil slick signatures in oceanographic SAR images. This method is based on directional behavioural oil slick in the sea surface. Therefore, directional hysteresis thresholding responses are first computed in order to bring to the fore dark spots, and increase pixels connexity in each Freeman direction. Those responses are then merged with the help of a Context Independent Constant Behaviour (CICB) operator. The proposed method is tested on ERS SAR amplitude images of Mediterranean and Atlantic seas. The obtained results are promising and show a good qualitative detection.

Independent Component Analysis (ICA) and Projection Pursuit (PP) are non orthogonal linear projection methods useful for dimensionality reduction of hyperspectral data cubes, in many cases more interesting than the standard Principal Component Analysis (PCA) but unfortunately not very robust to the noise. In this paper, the spatial correlation information is taken into account in order to improve the performances of both methods, following the ideas behind the so-called Noise-Adjusted Principal Component Analysis (NAPCA). This leads to the construction of two robust non orthogonal linear projection methods, respectively called Noise-Adjusted Independant Component Analysis (NAICA) and Noise-adjusted Projection Pursuit (NAPP).

The Hidden Markov Chain (HMC) model has been extended to take into consideration the multi-component representation of an hyperspectral data cube. Parameters estimation is performed using the general Iterative Conditional Estimation (ICE) method. The vectorial extension of the model is straightforward since the vectorial point of view joints the observation of each pixel as a spectral signature. Then, the segmentation procedure achieves an estimation of multi-dimensional correlated probability density functions (pdf). Multi-dimensional densities have been estimated by a set of 1D densities through a projection step that makes component independent and of reduced dimension. Classifications have been applied to an image from the CASI sensor including 17 bands (from 450 to 950 nm) representing an intensive agricultural region (Brittany, France). Since, the intrinsic dimensionality of the observation has been estimated to 4, the multi-component HMC model has been applied to the CASI image reduced to 4 bands through an adapted projection pursuit method.

A Markov chain model is applied for the segmentation of oil slicks acquired by SAR sensors. Actually, oil slicks have specific impact on ocean wave spectra. Initial wave spectra may be characterized by three kinds of waves, big, medium and small, which correspond physically to gravity and gravity-capillary waves. The increase of viscosity due to the presence of oil damps gravity-capillary waves. This induces a damping of the backscattering to the sensor, but also a dampening of the energy of the wave spectra. Thus, local segmentation of wave spectra may be achieved by the segmentation of a multiscale decomposition of the original SAR image. In this work, the unsupervised segmentation is achieved by using a vectorial extension of the Hidden Markov Chain (HMC) model. Parameters estimation is performed using the general Iterative Conditional Estimation (ICE) method. The problem of estimating multi-dimensional and non-Gaussian densities is solved by using a Principal Component Analysis (PCA). The algorithm has been applied on an ERS-PRI image. It yields interesting segmentation results with a very limited number of false alarms. Also, the multiscale segmentation proved to be an interesting alternative to classify marginal or degraded slicks.

Support vector machines (SVM) has been recently used with success for the classification of hyperspectral images. This method appears to be a robust alternative for pattern recognition with hyperspectral data: since the method is based on a geometric point of view, no statistical estimation has to be achieved. Then, SVM outperforms classical supervised classification algorithms such as the maximum likelihood when the number of spectral bands increases or when the number of training samples remains limited. Nevertheless, those kernel-based methods do not take into consideration the spectral similarity between support vectors. Then, some modified kernels are presented to take into consideration the spectral similarity between support vectors to outperform SVM-based classification of hyperspectral data cube. Those kernels (that still suit Mercer's conditions) are based on the use of spectral angle to evaluate the distance between support vectors. Classifiers to compare have been applied to an image from the CASI sensor including 17 bands from 450 to 950nm representing an intensive agricultural region (Brittany, France). It appears that those kernels reduce false alarms that were induced by illumination effects with classical kernels.

Synthetic aperture radar (SAR) images suffer from speckle noise that degrades image quality. Also, speckle removal is needed for lossy image compression as such noise increases image entropy. Nevertheless, most lossy compression schemes act as lowpass filters that might not suit multiplicative noise reduction. In order to build a compression scheme that takes into consideration statistics of SAR images, a well-known reflectivity estimator is integrated into an JPEG 2000-like image compression strategy, so that image compression artifacts may be viewed as an adaptive despeckle filter instead of a simple lowpass filter. It yields a compression algorithm that acts in a similar way to the Lee (or Kuan) filter.

Support vector machines, recently introduced in hyperspectral imagery, are applied to classify land cover on images from the airborne CASI sensor with a small training set. A smoothing preprocessing step is achieved, based on a vectorial extension of the anisotropic diffusion nonlinear filtering process. It allows the separability of the classes to be increased as well as homogeneous areas to be smoothed. It comes to take into consideration the spatial context before the classification, leading to improve the classification rate and to produce noiselessly classification maps with support vector machines.

A vectorial extension of the scalar anisotropic diffusion nonlinear filtering process applied on hyperspectral images is presented. In a first step, data are projected in a transformed space with a Maximum Noise Fraction transform, allowing the new components to be sorted in order of signal to noise ratio. The filtering is adapted to the signal to noise ratio of each component and a spectral dissimilarity vectorial measure is used in the filtering process. The inverse transform allows the filtered data to be reprojected in the original space. This process is useful for denoising hyperspectral images and for reducing spatial and spectral variability in each class of interest, leading to increase the performance of further segmentation or classification algorithms.

Many tools have been proposed in the literature for texture characterization of images. Some of them are based on statistical properties, others on fractal measures and some more on multiresolution analysis. Those methods have been proposed in a scalar point of view to be applied on mono-band images. They are not suited to the hyperspectral context where the spectral signature of each pixel has to be considered as a vector. Hyperspectral texture characterization is studied in this paper by extending the wavelet transform to suit hyperspectral images. A dimensionality reduction step is first applied on hyperspectral data with nonlinear transform, then a multiresolution analysis is performed on a limited number of spectral bands. The texture characterization by itself is based on the accurate modeling of the marginal distribution of the wavelet coefficients using Generalized Gaussian Density (GGD) and similarity measurement of GGDS by computing the Kullback-Leibler (KL) divergence. The results are presented with real hyperspectral images in a supervised and non-supervised texture segmentation.

In this paper, we propose a new method for compressing with loss temporal series of SAR images. The algorithm requires a reference scene used as an input in the compression process. The compression is achieved through affine transformations, comparable to the ones used in fractal coding, but applied here in a non-convergent way. Blocks in the image to be compressed are defined by a non-overlapping recursive partition. They are replaced by blocks taken from the reference image on which affine transformations, containing isometries and grey level adjustments, have been applied. This algorithm is very powerful, and outperforms classical algorithms (such that JPEG, EPIC or fractals) at high compression ratio with poor loss of critical information. Some results are presented with RADARSAT Antarctica and sea ice images which can be compressed at 44:1, and brought on icebreaker via an affordable INMARSAT connection, while still remaining of sufficient quality for ship routing usage.