Showing papers in "IEEE Transactions on Geoscience and Remote Sensing in 2007"
TL;DR: A detailed overview of the TanDEM-X mission concept is given which is based on the systematic combination of several innovative technologies, including a novel satellite formation flying concept allowing for the collection of bistatic data with short along-track baselines, as well as the use of new interferometric modes for system verification and DEM calibration.
Abstract: TanDEM-X (TerraSAR-X add-on for digital elevation measurements) is an innovative spaceborne radar interferometer that is based on two TerraSAR-X radar satellites flying in close formation. The primary objective of the TanDEM-X mission is the generation of a consistent global digital elevation model (DEM) with an unprecedented accuracy, which is equaling or surpassing the HRTI-3 specification. Beyond that, TanDEM-X provides a highly reconfigurable platform for the demonstration of new radar imaging techniques and applications. This paper gives a detailed overview of the TanDEM-X mission concept which is based on the systematic combination of several innovative technologies. The key elements are the bistatic data acquisition employing an innovative phase synchronization link, a novel satellite formation flying concept allowing for the collection of bistatic data with short along-track baselines, as well as the use of new interferometric modes for system verification and DEM calibration. The interferometric performance is analyzed in detail, taking into account the peculiarities of the bistatic operation. Based on this analysis, an optimized DEM data acquisition plan is derived which employs the combination of multiple data takes with different baselines. Finally, a collection of instructive examples illustrates the capabilities of TanDEM-X for the development and demonstration of new remote sensing applications.
1,235 citations
TL;DR: Multivariate regression is adopted to improve spectral quality, without diminishing spatial quality, in image fusion methods based on the well-established component substitution (CS) approach and quantitative scores carried out on spatially degraded data clearly confirm the superiority of the enhanced methods over their baselines.
Abstract: In this paper, multivariate regression is adopted to improve spectral quality, without diminishing spatial quality, in image fusion methods based on the well-established component substitution (CS) approach. A general scheme that is capable of modeling any CS image fusion method is presented and discussed. According to this scheme, a generalized intensity component is defined as the weighted average of the multispectral (MS) bands. The weights are obtained as regression coefficients between the MS bands and the spatially degraded panchromatic (Pan) image, with the aim of capturing the spectral responses of the sensors. Once it has been integrated into the Gram-Schmidt spectral-sharpening method, which is implemented in environment for visualizing images (ENVI) program, and into the generalized intensity-hue-saturation fusion method, the proposed preprocessing module allows the production of fused images of the same spatial sharpness but of increased spectral quality with respect to the standard implementations. In addition, quantitative scores carried out on spatially degraded data clearly confirm the superiority of the enhanced methods over their baselines.
895 citations
TL;DR: A novel method without the pure-pixel assumption is presented, referred to as the minimum volume constrained nonnegative matrix factorization (MVC-NMF), for unsupervised endmember extraction from highly mixed image data, which outperforms several other advanced endmember detection approaches.
Abstract: Endmember extraction is a process to identify the hidden pure source signals from the mixture. In the past decade, numerous algorithms have been proposed to perform this estimation. One commonly used assumption is the presence of pure pixels in the given image scene, which are detected to serve as endmembers. When such pixels are absent, the image is referred to as the highly mixed data, for which these algorithms at best can only return certain data points that are close to the real endmembers. To overcome this problem, we present a novel method without the pure-pixel assumption, referred to as the minimum volume constrained nonnegative matrix factorization (MVC-NMF), for unsupervised endmember extraction from highly mixed image data. Two important facts are exploited: First, the spectral data are nonnegative; second, the simplex volume determined by the endmembers is the minimum among all possible simplexes that circumscribe the data scatter space. The proposed method takes advantage of the fast convergence of NMF schemes, and at the same time eliminates the pure-pixel assumption. The experimental results based on a set of synthetic mixtures and a real image scene demonstrate that the proposed method outperforms several other advanced endmember detection approaches
870 citations
TL;DR: Two algorithms outperform all the others, the visual analysis being confirmed by the quantitative evaluation, and they basically rely on MRA and employ adaptive models for the injection of high-pass details.
Abstract: In January 2006, the Data Fusion Committee of the IEEE Geoscience and Remote Sensing Society launched a public contest for pansharpening algorithms, which aimed to identify the ones that perform best. Seven research groups worldwide participated in the contest, testing eight algorithms following different philosophies [component substitution, multiresolution analysis (MRA), detail injection, etc.]. Several complete data sets from two different sensors, namely, QuickBird and simulated Pleiades, were delivered to all participants. The fusion results were collected and evaluated, both visually and objectively. Quantitative results of pansharpening were possible owing to the availability of reference originals obtained either by simulating the data collected from the satellite sensor by means of higher resolution data from an airborne platform, in the case of the Pleiades data, or by first degrading all the available data to a coarser resolution and saving the original as the reference, in the case of the QuickBird data. The evaluation results were presented during the special session on data fusion at the 2006 international geoscience and remote sensing symposium in Denver, and these are discussed in further detail in this paper. Two algorithms outperform all the others, the visual analysis being confirmed by the quantitative evaluation. These two methods share the same philosophy: they basically rely on MRA and employ adaptive models for the injection of high-pass details.
789 citations
TL;DR: The proposed technique would also allow precise coregistration of images for the measurement of surface displacements due to ice-flow or geomorphic processes, or for any other change detection applications.
Abstract: We describe a procedure to accurately measure ground deformations from optical satellite images. Precise orthorectification is obtained owing to an optimized model of the imaging system, where look directions are linearly corrected to compensate for attitude drifts, and sensor orientation uncertainties are accounted for. We introduce a new computation of the inverse projection matrices for which a rigorous resampling is proposed. The irregular resampling problem is explicitly addressed to avoid introducing aliasing in the ortho-rectified images. Image registration and correlation is achieved with a new iterative unbiased processor that estimates the phase plane in the Fourier domain for subpixel shift detection. Without using supplementary data, raw images are wrapped onto the digital elevation model and coregistered with a 1/50 pixel accuracy. The procedure applies to images from any pushbroom imaging system. We analyze its performance using Satellite pour l'Observation de la Terre (SPOT) images in the case of a null test (no coseismic deformation) and in the case of large coseismic deformations due to the Mw 7.1 Hector Mine, California, earthquake of 1999. The proposed technique would also allow precise coregistration of images for the measurement of surface displacements due to ice-flow or geomorphic processes, or for any other change detection applications. A complete software package, the Coregistration of Optically Sensed Images and Correlation, is available for download from the Caltech Tectonics Observatory website
777 citations
TL;DR: The introduction of the composite-kernel framework drastically improves results, and the new fast formulation ranks almost linearly in the computational cost, rather than cubic as in the original method, thus allowing the use of this method in remote-sensing applications.
Abstract: This paper presents a semi-supervised graph-based method for the classification of hyperspectral images. The method is designed to handle the special characteristics of hyperspectral images, namely, high-input dimension of pixels, low number of labeled samples, and spatial variability of the spectral signature. To alleviate these problems, the method incorporates three ingredients, respectively. First, being a kernel-based method, it combats the curse of dimensionality efficiently. Second, following a semi-supervised approach, it exploits the wealth of unlabeled samples in the image, and naturally gives relative importance to the labeled ones through a graph-based methodology. Finally, it incorporates contextual information through a full family of composite kernels. Noting that the graph method relies on inverting a huge kernel matrix formed by both labeled and unlabeled samples, we originally introduce the Nystro umlm method in the formulation to speed up the classification process. The presented semi-supervised-graph-based method is compared to state-of-the-art support vector machines in the classification of hyperspectral data. The proposed method produces better classification maps, which capture the intrinsic structure collectively revealed by labeled and unlabeled points. Good and stable accuracy is produced in ill-posed classification problems (high dimensional spaces and low number of labeled samples). In addition, the introduction of the composite-kernel framework drastically improves results, and the new fast formulation ranks almost linearly in the computational cost, rather than cubic as in the original method, thus allowing the use of this method in remote-sensing applications.
589 citations
TL;DR: A new similarity measure for automatic change detection in multitemporal synthetic aperture radar images based on the evolution of the local statistics of the image between two dates, which allows a multiscale approach in the change detection for operational use.
Abstract: 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
500 citations
TL;DR: This paper addresses unsupervised change detection by proposing a proper framework for a formal definition and a theoretical study of the change vector analysis (CVA) technique and the results obtained confirm the interest of the proposed framework and the validity of the related theoretical analysis.
Abstract: This paper addresses unsupervised change detection by proposing a proper framework for a formal definition and a theoretical study of the change vector analysis (CVA) technique. This framework, which is based on the representation of the CVA in polar coordinates, aims at: 1) introducing a set of formal definitions in the polar domain (which are linked to the properties of the data) for a better general description (and thus understanding) of the information present in spectral change vectors; 2) analyzing from a theoretical point of view the distributions of changed and unchanged pixels in the polar domain (also according to possible simplifying assumptions); 3) driving the implementation of proper preprocessing procedures to be applied to multitemporal images on the basis of the results of the theoretical study on the distributions; and 4) defining a solid background for the development of advanced and accurate automatic change-detection algorithms in the polar domain. The findings derived from the theoretical analysis on the statistical models of classes have been validated on real multispectral and multitemporal remote sensing images according to both qualitative and quantitative analyses. The results obtained confirm the interest of the proposed framework and the validity of the related theoretical analysis
486 citations
TL;DR: This paper presents a technique for dimensionality reduction to deal with hyperspectral images based on a hierarchical clustering structure to group bands to minimize the intracluster variance and maximize the intercluster variance.
Abstract: Hyperspectral imaging involves large amounts of information. This paper presents a technique for dimensionality reduction to deal with hyperspectral images. The proposed method is based on a hierarchical clustering structure to group bands to minimize the intracluster variance and maximize the intercluster variance. This aim is pursued using information measures, such as distances based on mutual information or Kullback-Leibler divergence, in order to reduce data redundancy and non useful information among image bands. Experimental results include a comparison among some relevant and recent methods for hyperspectral band selection using no labeled information, showing their performance with regard to pixel image classification tasks. The technique that is presented has a stable behavior for different image data sets and a noticeable accuracy, mainly when selecting small sets of bands.
477 citations
TL;DR: The PALSAR observation strategy has been designed to provide consistent, wall-to-wall observations at fine resolution of all land areas on the Earth on a repetitive basis, in a manner which has earlier been conceived only for coarse- and medium-resolution instruments.
Abstract: The Advanced Land Observing Satellite (ALOS) is Japan's new-generation Earth Observation satellite, launched in January 2006 by the Japan Aerospace Exploration Agency. ALOS carries two optical instruments (Panchromatic Remote-sensing Instrument for Stereo Mapping and Advanced Visible and Near-Infrared Radiometer type 2) and, to maintain Japan's commitment to spaceborne L-band Synthetic Aperture Radar (SAR), the Phased Array L-band SAR (PALSAR). The successor to the SAR onboard the Japanese Earth Resources Satellite (1992-1998), the PALSAR instrument provides enhanced sensor characteristics, including full polarimetry, variable off-nadir viewing, and ScanSAR operations, as well as significantly improved radiometric and geometric performance. As important as the technical improvements and the reason PALSAR here is referred to as a pathfinder mission for global environmental monitoring is the systematic data-acquisition strategy which has been implemented for ALOS. With a priority second only to emergency observations, the PALSAR observation strategy has been designed to provide consistent, wall-to-wall observations at fine resolution of all land areas on the Earth on a repetitive basis, in a manner which has earlier been conceived only for coarse- and medium-resolution instruments.
471 citations
TL;DR: It is shown that both the phase and magnitude of the symmetric scattering type should be used for an unambiguous description of symmetric target scattering.
Abstract: The Kennaugh-Huynen scattering matrix con-diagonalization is projected into the Pauli basis to derive a new scattering vector model for the representation of coherent target scattering. This model permits a polarization basis invariant representation of coherent target scattering in terms of five independent target parameters, the magnitude and phase of the symmetric scattering type introduced in this paper, and the maximum polarization parameters (orientation, helicity, and maximum return). The new scattering vector model served for the assessment of the Cloude-Pottier incoherent target decomposition. Whereas the Cloude-Pottier scattering type alpha and entropy H are roll invariant, beta and the so-called target-phase parameters do depend on the target orientation angle for asymmetric scattering. The scattering vector model is then used as the basis for the development of new coherent and incoherent target decompositions in terms of unique and roll-invariant target parameters. It is shown that both the phase and magnitude of the symmetric scattering type should be used for an unambiguous description of symmetric target scattering. Target helicity is required for the assessment of the symmetry-asymmetry nature of target scattering. The symmetric scattering type phase is shown to be very promising for wetland classification in particular, using polarimetric Convair-580 synthetic aperture radar data collected over the Ramsar Mer Bleue wetland site to the east of Ottawa, Ontario, Canada
TL;DR: The proposed SVM-based fusion approach outperforms all other approaches and significantly improves the results of a single SVM, which is trained on the whole multisensor data set.
Abstract: The classification of multisensor data sets, consisting of multitemporal synthetic aperture radar data and optical imagery, is addressed. The concept is based on the decision fusion of different outputs. Each data source is treated separately and classified by a support vector machine (SVM). Instead of fusing the final classification outputs (i.e., land cover classes), the original outputs of each SVM discriminant function are used in the subsequent fusion process. This fusion is performed by another SVM, which is trained on the a priori outputs. In addition, two voting schemes are applied to create the final classification results. The results are compared with well-known parametric and nonparametric classifier methods, i.e., decision trees, the maximum-likelihood classifier, and classifier ensembles. The proposed SVM-based fusion approach outperforms all other approaches and significantly improves the results of a single SVM, which is trained on the whole multisensor data set.
TL;DR: The aim of the experiment was to demonstrate that interferometric synthetic aperture radar (InSAR) measurements can indeed allow a displacement time series estimation with submillimeter accuracy (both in horizontal and vertical directions), provided that the data are properly processed and the impact of in situ as well as atmospheric effects is minimized.
Abstract: This paper presents the results of a blind experiment that is performed using two pairs of dihedral reflectors. The aim of the experiment was to demonstrate that interferometric synthetic aperture radar (InSAR) measurements can indeed allow a displacement time series estimation with submillimeter accuracy (both in horizontal and vertical directions), provided that the data are properly processed and the impact of in situ as well as atmospheric effects is minimized. One pair of dihedral reflectors was moved a few millimeters between SAR acquisitions, in the vertical and east-west (EW) directions, and the ground truth was compared with the InSAR data. The experiment was designed to allow a multiplatform and multigeometry analysis, i.e., each reflector was carefully pointed in order to be visible in both Envisat and Radarsat acquisitions. Moreover, two pairs of reflectors were used to allow the combination of data gathered along ascending and descending orbits. The standard deviation of the error is 0.75 mm in the vertical direction and 0.58 mm in the horizontal (EW) direction. GPS data were also collected during this experiment in order to cross-check the SAR results
TL;DR: A novel processing solution is presented, which solves the nonlinearity problem for the whole range profile of the FMCW SAR signal model and is applied to stripmap, spotlight, and single-transmitter/multiple-receiver digital-beamforming SAR operational mode.
Abstract: The combination of frequency-modulated continuous-wave (FMCW) technology and synthetic aperture radar (SAR) techniques leads to lightweight cost-effective imaging sensors of high resolution. One limiting factor to the use of FMCW sensors is the well-known presence of nonlinearities in the transmitted signal. This results in contrast- and range-resolution degradation, particularly when the system is intended for high-resolution long-range applications, as it is the case for SAR. This paper presents a novel processing solution, which solves the nonlinearity problem for the whole range profile. Additionally, the conventional stop-and-go approximation used in pulse-radar algorithms is not valid in FMCW SAR applications under certain circumstances. Therefore, the motion within the sweep needs to be taken into account. Analytical development of the FMCW SAR signal model, starting from the deramped signal and without using the stop-and-go approximation, is presented in this paper. The model is then applied to stripmap, spotlight, and single-transmitter/multiple-receiver digital-beamforming SAR operational mode. The proposed algorithms are verified by processing real FMCW SAR data collected with the demonstrator system built at the Delft University of Technology.
TL;DR: An automatic road seeding method based on rectangular approximations to road footprints and a toe-finding algorithm to classify footprints for growing a road tree and introduces a lognormal distribution to characterize the conditional probability of A/P ratios of the footprints in the road tree.
Abstract: In this paper, a new two-step approach (detecting and pruning) for automatic extraction of road networks from aerial images is presented. The road detection step is based on shape classification of a local homogeneous region around a pixel. The local homogeneous region is enclosed by a polygon, called the footprint of the pixel. This step involves detecting road footprints, tracking roads, and growing a road tree. We use a spoke wheel operator to obtain the road footprint. We propose an automatic road seeding method based on rectangular approximations to road footprints and a toe-finding algorithm to classify footprints for growing a road tree. The road tree pruning step makes use of a Bayes decision model based on the area-to-perimeter ratio (the A/P ratio) of the footprint to prune the paths that leak into the surroundings. We introduce a lognormal distribution to characterize the conditional probability of A/P ratios of the footprints in the road tree and present an automatic method to estimate the parameters that are related to the Bayes decision model. Results are presented for various aerial images. Evaluation of the extracted road networks using representative aerial images shows that the completeness of our road tracker ranges from 84% to 94%, correctness is above 81%, and quality is from 82% to 92%.
TL;DR: This paper describes successful rapid satellite mapping campaigns supporting disaster relief and demonstrates how this technology can be used for civilian crisis-management purposes and reports on rapid-mapping experiences gained during various disaster-response applications.
Abstract: This paper describes how multisource satellite data and efficient image analysis may successfully be used to conduct rapid-mapping tasks in the domain of disaster and crisis-management support. The German Aerospace Center (DLR) has set up a dedicated crosscutting service, which is the so-called "Center for satellite-based Crisis Information" (ZKI), to facilitate the use of its Earth-observation capacities in the service of national and international response to major disaster situations, humanitarian relief efforts, and civil security issues. This paper describes successful rapid satellite mapping campaigns supporting disaster relief and demonstrates how this technology can be used for civilian crisis-management purposes. During the last years, various international coordination bodies were established, improving the disaster-response-related cooperation within the Earth-observation community worldwide. DLR/ZKI operates in this context, closely networking with public authorities (civil security), nongovernmental organizations (humanitarian relief organizations), satellite operators, and other space agencies. This paper reflects on several of these international activities, such as the International Charter Space and Major Disasters, describes mapping procedures, and reports on rapid-mapping experiences gained during various disaster-response applications. The example cases presented cover rapid impact assessment after the Indian Ocean Tsunami, forest fires mapping for Portugal, earthquake-damage assessment for Pakistan, and landslide extent mapping for the Philippines
TL;DR: The MWRRET algorithm significantly provides more accurate retrievals than the original ARM statistical retrieval, which uses monthly retrieval coefficients, by combining the two retrieval methods with the application of brightness temperature offsets to reduce the spurious LWP bias in clear skies.
Abstract: Ground-based two-channel microwave radiometers (MWRs) have been used for over 15 years by the Atmospheric Radiation Measurement (ARM) program to provide observations of downwelling emitted radiance from which precipitable water vapor (PWV) and liquid water path (LWP) - two geophysical parameters critical for many areas of atmospheric research - are retrieved. An algorithm that incorporates output from two advanced retrieval techniques, namely, a physical-iterative approach and a computationally efficient statistical method, has been developed to retrieve these parameters. The forward model used in both methods is the monochromatic radiative transfer model MonoRTM. An important component of this MWR RETrieval (MWRRET) algorithm is the determination of small (< 1 K) offsets that are subtracted from the observed brightness temperatures before the retrievals are performed. Accounting for these offsets removes systematic biases from the observations and/or the model spectroscopy necessary for the retrieval, significantly reducing the systematic biases in the retrieved LWP. The MWRRET algorithm significantly provides more accurate retrievals than the original ARM statistical retrieval, which uses monthly retrieval coefficients. By combining the two retrieval methods with the application of brightness temperature offsets to reduce the spurious LWP bias in clear skies, the MWRRET algorithm significantly provides better retrievals of PWV and LWP from the ARM two-channel MWRs compared to the original ARM product.
TL;DR: The primary science objective of this mission is to monitor the seasonal and interannual variation of the large-scale features of the surface salinity field in the open ocean with a spatial resolution of 150 km and a retrieval accuracy of 0.2 psu globally on a monthly basis.
Abstract: Aquarius is a combined passive/active L-band microwave instrument that is being developed to map the salinity field at the surface of the ocean from space. The data will support studies of the coupling between ocean circulation, global water cycle, and climate. Aquarius is part of the Aquarius/Satelite de Aplicaciones Cientiflcas-D mission, which is a partnership between the U.S. (National Aeronautics and Space Administration) and Argentina (Comision Nacional de Actividades Espaciales). The primary science objective of this mission is to monitor the seasonal and interannual variation of the large-scale features of the surface salinity field in the open ocean with a spatial resolution of 150 km and a retrieval accuracy of 0.2 psu globally on a monthly basis.
TL;DR: Results demonstrate that the MCC model minimizes commission errors while retaining a high proportion of ground returns and provides high confidence in the derived ground surface.
Abstract: One prerequisite to the use of light detection and ranging (LiDAR) across disciplines is differentiating ground from nonground returns. The objective was to automatically and objectively classify points within unclassified LiDAR point clouds, with few model parameters and minimal postprocessing. Presented is an automated method for classifying LiDAR returns as ground or nonground in forested environments occurring in complex terrains. Multiscale curvature classification (MCC) is an iterative multiscale algorithm for classifying LiDAR returns that exceed positive surface curvature thresholds, resulting in all the LiDAR measurements being classified as ground or nonground. The MCC algorithm yields a solution of classified returns that support bare-earth surface interpolation at a resolution commensurate with the sampling frequency of the LiDAR survey. Errors in classified ground returns were assessed using 204 independent validation points consisting of 165 field plot global positioning system locations and 39 National Oceanic and Atmospheric Administration-National Geodetic Survey monuments. Jackknife validation and Monte Carlo simulation were used to assess the quality and error of a bare-earth digital elevation model interpolated from the classified returns. A local indicator of spatial association statistic was used to test for commission errors in the classified ground returns. Results demonstrate that the MCC model minimizes commission errors while retaining a high proportion of ground returns and provides high confidence in the derived ground surface
TL;DR: It is shown that the scheme based on the proposed low-complexity KLT significantly outperforms previous schemes as to rate-distortion performance, and an evaluation framework based on both reconstruction fidelity and impact on image exploitation is introduced.
Abstract: Transform-based lossy compression has a huge potential for hyperspectral data reduction. Hyperspectral data are 3-D, and the nature of their correlation is different in each dimension. This calls for a careful design of the 3-D transform to be used for compression. In this paper, we investigate the transform design and rate allocation stage for lossy compression of hyperspectral data. First, we select a set of 3-D transforms, obtained by combining in various ways wavelets, wavelet packets, the discrete cosine transform, and the Karhunen-Loegraveve transform (KLT), and evaluate the coding efficiency of these combinations. Second, we propose a low-complexity version of the KLT, in which complexity and performance can be balanced in a scalable way, allowing one to design the transform that better matches a specific application. Third, we integrate this, as well as other existing transforms, in the framework of Part 2 of the Joint Photographic Experts Group (JPEG) 2000 standard, taking advantage of the high coding efficiency of JPEG 2000, and exploiting the interoperability of an international standard. We introduce an evaluation framework based on both reconstruction fidelity and impact on image exploitation, and evaluate the proposed algorithm by applying this framework to AVIRIS scenes. It is shown that the scheme based on the proposed low-complexity KLT significantly outperforms previous schemes as to rate-distortion performance. As for impact on exploitation, we consider multiclass hard classification, spectral unmixing, binary classification, and anomaly detection as benchmark applications
TL;DR: A multiobjective optimization algorithm is utilized to tackle the problem of fuzzy partitioning where a number of fuzzy cluster validity indexes are simultaneously optimized and the resultant set of near-Pareto-optimal solutions contains aNumber of nondominated solutions, which the user can judge relatively and pick up the most promising one according to the problem requirements.
Abstract: An important approach for unsupervised landcover classification in remote sensing images is the clustering of pixels in the spectral domain into several fuzzy partitions. In this paper, a multiobjective optimization algorithm is utilized to tackle the problem of fuzzy partitioning where a number of fuzzy cluster validity indexes are simultaneously optimized. The resultant set of near-Pareto-optimal solutions contains a number of nondominated solutions, which the user can judge relatively and pick up the most promising one according to the problem requirements. Real-coded encoding of the cluster centers is used for this purpose. Results demonstrating the effectiveness of the proposed technique are provided for numeric remote sensing data described in terms of feature vectors. Different landcover regions in remote sensing imagery have also been classified using the proposed technique to establish its efficiency
TL;DR: Algorithms for automatic detection of oil spills in synthetic aperture radar (SAR) images are presented and can be considered to be a good alternative to manual inspection when large ocean areas are to be inspected.
Abstract: We present algorithms for automatic detection of oil spills in synthetic aperture radar (SAR) images. The algorithms consist of three main parts, namely: 1) detection of dark spots; 2) feature extraction from the dark spot candidates; and 3) classification of dark spots as oil spills or look-alikes. The algorithms have been trained on a large number of Radarsat and Envisat Advanced Synthetic Aperture Radar (ASAR) images. The performance of the algorithm is compared to manual and semiautomatic approaches in a benchmark study using 59 Radarsat and Envisat images. The algorithms can be considered to be a good alternative to manual inspection when large ocean areas are to be inspected
TL;DR: The four-stream radiative transfer (RT) formalism of the Scattering by Arbitrarily Inclined Leaves model family is extended to the TIR domain to simulate the multiple scattering and emission inside a geometrically homogenous but thermodynamically heterogeneous canopy for optical as well as thermal radiation using the same modeling framework.
Abstract: Foliage and soil temperatures are key variables for assessing the exchanges of turbulent heat fluxes between vegetated land and the atmosphere. Using multiple-view-angle thermal-infrared (TIR) observations, the temperatures of soil and vegetation may be retrieved. However, particularly for sparsely vegetated areas, the soil and vegetation component temperatures in the sun and in the shade may be very different depending on the solar radiation, the physical properties of the surface, and the meteorological conditions. This may interfere with a correct retrieval of component temperatures, but it might also yield extra information related to canopy structure. Both are strong reasons to investigate this phenomenon in some more detail. To this end, the relationship between the TIR radiance directionality and the component temperatures has been analyzed. In this paper, we extend the four-stream radiative transfer (RT) formalism of the Scattering by Arbitrarily Inclined Leaves model family to the TIR domain. This new approach enables us to simulate the multiple scattering and emission inside a geometrically homogenous but thermodynamically heterogeneous canopy for optical as well as thermal radiation using the same modeling framework. In this way top-of-canopy thermal radiances observed under multiple viewing angles can be related to the temperatures of sunlit and shaded soil and sunlit and shaded leaves. In this paper, we describe the development of this unified optical-thermal RT theory and demonstrate its capabilities. A preliminary validation using an experimental data set collected in the Shunyi remote sensing field campaign in China is briefly addressed
TL;DR: Experimental results that are obtained on multitemporal RADARSAT-1 SAR images of the Sumatra Island, Indonesia, confirm the effectiveness of both the proposed SBA and the presented system for tsunami-damage assessment.
Abstract: This paper presents a split-based approach (SBA) to automatic and unsupervised change detection in large-size multitemporal remote-sensing images. Unlike standard methods that are presented in the literature, the proposed approach can detect in a consistent and reliable way changes in images of large size also when the extension of the changed area is small (and, therefore, the prior probability of the class of changed pixels is very small). The method is based on the following: 1) a split of the large-size image into subimages; 2) an adaptive analysis of each subimage; and 3) an automatic split-based threshold-selection procedure. This general approach is used for defining a system for damage assessment in multitemporal synthetic aperture radar (SAR) images. The proposed system has been developed to properly identify different levels of damages that are induced by tsunamis along coastal areas. Experimental results that are obtained on multitemporal RADARSAT-1 SAR images of the Sumatra Island, Indonesia, confirm the effectiveness of both the proposed SBA and the presented system for tsunami-damage assessment
TL;DR: Results from the SDSM on-orbit observations show that the SD bidirectional reflectance factor (BRF) also has a similar wavelength-dependent degradation, with the largest degradation appearing at the shortest wavelengths.
Abstract: Terra Moderate Resolution Imaging Spectroradiometer (MODIS) has made continuous global observations for more than six years since its launch in December 1999. MODIS has 36 spectral bands: 20 reflective solar bands (RSBs) with wavelengths from 0.41-2.2 mum and 16 thermal emissive bands with wavelengths from 3.7-14.4 mum. It is a cross-track scanning radiometer that collects data at three nadir spatial resolutions: 0.25 km (2 bands), 0.5 km (5 bands), and 1 km (29 bands). An onboard solar diffuser (SD) and an SD stability monitor (SDSM) are used biweekly for RSB on-orbit radiometric calibration. Another onboard calibrator (OBC), a spectroradiometric calibration assembly, is used periodically to evaluate and monitor RSB spatial and spectral performance. In addition to measurements made using OBCs, lunar observations at nearly identical phase angles are used to track RSB calibration stability. This paper provides an overview of MODIS RSB on-orbit calibration algorithms and operational activities. It discusses sensor characteristics that could impact RSB calibration accuracy and data product quality, including degradation of the SD bidirectional reflectance factor (BRF), degradation of the scan mirror reflectance in the visible spectral region, and changes in operational configuration. The Terra MODIS OBCs have performed well in monitoring SD degradation and tracking changes in RSB response. Band 8 (0.41 mum) has experienced the largest response decrease with an approximate annual rate of 4.5% (mirror side 1). Band 9 (0.44 mum) has an annual response decrease of about 2.3% (mirror side 1). For most RSB bands with wavelengths greater than 0.5 mum, the annual response changes are generally less than 1.0%. Results from the SDSM on-orbit observations show that the SD BRF also has a similar wavelength-dependent degradation, with the largest degradation appearing at the shortest wavelengths. Among the 330 RSB detectors, there are no inoperable detectors, and only a few noisy detectors have appeared postlaunch
TL;DR: Results show that the new polarimetric approach is able to assist classification, and the target decomposition theorem is exploited to distinguish oil spills and look-alikes.
Abstract: A study on sea oil spill observation by means of polarimetric synthetic aperture radar (SAR) data is accomplished. It is based on the use of a polarimetric constant false alarm rate filter to detect dark patches over SAR images. Then, the target decomposition theorem is exploited to distinguish oil spills and look-alikes. Experiments are conducted on polarimetric SAR data acquired during the SIR-C/X-SAR mission on October 1994. The data were processed and calibrated at the Jet Propulsion Laboratory, National Aeronautics and Space Administration. Results show that the new polarimetric approach is able to assist classification
TL;DR: The results indicate that the registration accuracy of ARRSI is comparable to that produced by a human expert and improvement over the baseline and multimodal sum of squared differences registration techniques tested.
Abstract: This paper presents the Automatic Registration of Remote-Sensing Images (ARRSI); an automatic registration system built to register satellite and aerial remotely sensed images. The system is designed specifically to address the problems associated with the registration of remotely sensed images obtained at different times and/or from different sensors. The ARRSI system is capable of handling remotely sensed images geometrically distorted by various transformations such as translation, rotation, and shear. Global and local contrast issues associated with remotely sensed images are addressed in ARRSI using control-point detection and matching processes based on a phase-congruency model. Intensity-difference issues associated with multimodal registration of remotely sensed images are addressed in ARRSI through the use of features that are invariant to intensity mappings during the control-point matching process. An adaptive control-point matching scheme is employed in ARRSI to reduce the performance issues associated with the registration of large remotely sensed images. Finally, a variation on the Random Sample and Consensus algorithm called Maximum Distance Sample Consensus is introduced in ARRSI to improve the accuracy of the transformation model between two remotely sensed images while minimizing computational overhead. The ARRSI system has been tested using various satellite and aerial remotely sensed images and evaluated based on its accuracy and computational performance. The results indicate that the registration accuracy of ARRSI is comparable to that produced by a human expert and improvement over the baseline and multimodal sum of squared differences registration techniques tested
TL;DR: Time-dependent response versus scan angle (RVS) lookup tables derived from lunar views, SD calibration, and Earth-view observations have been used to maintain the quality of the L1B data for both the Terra and Aqua MODIS RSB.
Abstract: The moderate resolution imaging spectroradiometer (MODIS) protoflight model on-board the Terra spacecraft and the MODIS flight model 1 on-board the Aqua spacecraft were launched on December 18, 1999 and May 4, 2002, respectively. They view the moon through the space view (SV) port approximately once a month to monitor the long-term radiometric stability of their reflective solar bands (RSBs). The lunar irradiance observed by MODIS depends on the viewing geometry. Algorithms were developed to select lunar views such that these geometric effects are minimized. In each MODIS lunar observation, the moon can be viewed in multiple scans. The lunar irradiance of a MODIS RSB can be derived from the response of all detectors of a spectral band in one scan which fully covers the moon, from that of one detector in multiple scans or from the response of all detectors in multiple scans. Based on lunar observations, a set of coefficients is defined and derived to trend MODIS system response degradation at the angle of incidence (AOI) of its SV port. It is shown that the degradation is both wavelength and mirror side dependent. Since launch, Terra and Aqua MODIS band 8 (412 nm) mirror side one have degraded 36% and 17%, respectively, at the AOI of the SV. A comparison between the lunar coefficients and those derived from the MODIS on-board solar diffuser (SD) calibrations shows that the response change of the MODIS RSB is both AOI and time dependent. Time-dependent response versus scan angle (RVS) lookup tables derived from lunar views, SD calibration, and Earth-view observations have been used to maintain the quality of the L1B data for both the Terra and Aqua MODIS RSB. The corrections provided by the RVS in the Terra and Aqua MODIS data from the 412-nm band are as large as 14% and 6.2%, respectively.
TL;DR: This paper presents the recently developed content-based multimodal Geospatial Information Retrieval and Indexing System (GeoIRIS) which includes automatic feature extraction, visual content mining from large-scale image databases, and high-dimensional database indexing for fast retrieval.
Abstract: Searching for relevant knowledge across heterogeneous geospatial databases requires an extensive knowledge of the semantic meaning of images, a keen eye for visual patterns, and efficient strategies for collecting and analyzing data with minimal human intervention. In this paper, we present our recently developed content-based multimodal Geospatial Information Retrieval and Indexing System (GeoIRIS) which includes automatic feature extraction, visual content mining from large-scale image databases, and high-dimensional database indexing for fast retrieval. Using these underpinnings, we have developed techniques for complex queries that merge information from heterogeneous geospatial databases, retrievals of objects based on shape and visual characteristics, analysis of multiobject relationships for the retrieval of objects in specific spatial configurations, and semantic models to link low-level image features with high-level visual descriptors. GeoIRIS brings this diverse set of technologies together into a coherent system with an aim of allowing image analysts to more rapidly identify relevant imagery. GeoIRIS is able to answer analysts' questions in seconds, such as "given a query image, show me database satellite images that have similar objects and spatial relationship that are within a certain radius of a landmark."
TL;DR: A full-waveform inversion scheme that is based on a finite-difference time-domain solution of Maxwell's equations is introduced and is shown to be remarkably robust to the presence of uncorrelated noise in the radar data.
Abstract: Crosshole radar techniques are important tools for a wide range of geoscientific and engineering investigations. Unfortunately, the resolution of crosshole radar images may be limited by inadequacies of the ray tomographic methods that are commonly used in inverting the data. Since ray methods are based on high-frequency approximations and only account for a small fraction of the information contained in the radar traces, they are restricted to resolving relatively large-scale features. As a consequence, the true potential of crosshole radar techniques has yet to be realized. To address this issue, we introduce a full-waveform inversion scheme that is based on a finite-difference time-domain solution of Maxwell's equations. We benchmark our new scheme on synthetic crosshole data generated from suites of increasingly complex models. The full-waveform tomographic images accurately reconstruct the following: (1) the locations, sizes, and electrical properties of isolated subwavelength objects embedded in homogeneous media; (2) the locations and sizes of adjacent subwavelength objects embedded in homogeneous media; (3) abrupt media boundaries and average and stochastic electrical property variations of heterogeneous layered models; and (4) the locations, sizes, and electrical conductivities of water-filled tunnels and closely spaced subwavelength pipes embedded in heterogeneous layered models. The new scheme is shown to be remarkably robust to the presence of uncorrelated noise in the radar data. Several limitations of the full-waveform tomographic inversion are also identified. For typical crosshole acquisition geometries and parameters, small resistive bodies and small closely spaced dielectric objects may be difficult to resolve. Furthermore, electrical property contrasts may be underestimated. Nevertheless, the full-waveform inversions usually provide substantially better results than those supplied by traditional ray methods.