Showing papers in "IEEE Geoscience and Remote Sensing Letters in 2011"
TL;DR: A technique based on independent component analysis (ICA) and extended morphological attribute profiles (EAPs) is presented for the classification of hyperspectral images and the effectiveness of the proposed technique was proved.
Abstract: In this letter, a technique based on independent component analysis (ICA) and extended morphological attribute profiles (EAPs) is presented for the classification of hyperspectral images. The ICA maps the data into a subspace in which the components are as independent as possible. APs, which are extracted by using several attributes, are applied to each image associated with an extracted independent component, leading to a set of extended EAPs. Two approaches are presented for including the computed profiles in the analysis. The features extracted by the morphological processing are then classified with an SVM. The experiments carried out on two hyperspectral images proved the effectiveness of the proposed technique.
405 citations
TL;DR: A pioneered mini-UAV-borne LIDAR system - Sensei is established schematically with an Ibeo Lux scanner mounted on a small Align T-Rex 600E helicopter to validate its applicability for fine-scale mapping, in terms of, e.g., tree height estimation, pole detection, road extraction, and digital terrain model refinement.
Abstract: Light detection and ranging (LIDAR) systems based on unmanned aerial vehicles (UAVs) recently are in rapid advancement, while mini-UAV-borne laser scanning has few reported progress, notwithstanding so extensively required. This study established a pioneered mini-UAV-borne LIDAR system - Sensei, schematically with an Ibeo Lux scanner mounted on a small Align T-Rex 600E helicopter. Furthermore, the associated data processing involved in the coordinate triple, pulse intensity, and multiechoes per pulse was explored to validate its applicability for fine-scale mapping, in terms of, e.g., tree height estimation, pole detection, road extraction, and digital terrain model refinement. The feasibility and advantages of mini-UAV-borne LIDAR have been demonstrated by the promising results based on the real-measured data.
292 citations
TL;DR: A new supervised band-selection algorithm that uses the known class signatures only without examining the original bands or the need of class training samples is proposed, which can complete the task much faster than traditional methods that test bands or band combinations.
Abstract: Band selection is often applied to reduce the dimensionality of hyperspectral imagery. When the desired object information is known, it can be achieved by finding the bands that contain the most object information. It is expected that these bands can provide an overall satisfactory detection and classification performance. In this letter, we propose a new supervised band-selection algorithm that uses the known class signatures only without examining the original bands or the need of class training samples. Thus, it can complete the task much faster than traditional methods that test bands or band combinations. The experimental result shows that our approach can generally yield better results than other popular supervised band-selection methods in the literature.
249 citations
TL;DR: The achieved improvement with respect to the single-output regression approach suggests that M-SVR can be considered a convenient alternative for nonparametric biophysical parameter estimation and model inversion.
Abstract: This letter proposes a multioutput support vector regression (M-SVR) method for the simultaneous estimation of different biophysical parameters from remote sensing images. General retrieval problems require multioutput (and potentially nonlinear) regression methods. M-SVR extends the single-output SVR to multiple outputs maintaining the advantages of a sparse and compact solution by using an e-insensitive cost function. The proposed M-SVR is evaluated in the estimation of chlorophyll content, leaf area index and fractional vegetation cover from a hyperspectral compact high-resolution imaging spectrometer images. The achieved improvement with respect to the single-output regression approach suggests that M-SVR can be considered a convenient alternative for nonparametric biophysical parameter estimation and model inversion.
242 citations
TL;DR: A semi-automatic approach based on object-oriented change detection for landslide rapid mapping and using very high resolution optical images is introduced, demonstrating the usefulness of this methodology on the Messina landslide event in southern Italy.
Abstract: A complete multitemporal landslide inventory, ideally updated after each major event, is essential for quantitative landslide hazard assessment. However, traditional mapping methods, which rely on manual interpretation of aerial photographs and intensive field surveys, are time consuming and not efficient for generating such event-based inventories. In this letter, a semi-automatic approach based on object-oriented change detection for landslide rapid mapping and using very high resolution optical images is introduced. The usefulness of this methodology is demonstrated on the Messina landslide event in southern Italy that occurred on October 1, 2009. The algorithm was first developed in a training area of Altolia and subsequently tested without modifications in an independent area of Itala. Correctly detected were 198 newly triggered landslides, with user accuracies of 81.8% for the number of landslides and 75.9% for the extent of landslides. The principal novelties of this letter are as follows: 1) a fully automatic problem-specified multiscale optimization for image segmentation and 2) a multitemporal analysis at object level with several systemized spectral and textural measurements.
240 citations
TL;DR: This letter presents a method for satellites image classification involving visual attention into the satellite image classification; biologically inspired saliency information is exploited in the phase of the image representation, making the method more concentrated on the interesting objects and structures.
Abstract: This letter presents a method for satellite image classification aiming at the following two objectives: 1) involving visual attention into the satellite image classification; biologically inspired saliency information is exploited in the phase of the image representation, making our method more concentrated on the interesting objects and structures, and 2) handling the satellite image classification without the learning phase. A two-layer sparse coding (TSC) model is designed to discover the “true” neighbors of the images and bypass the intensive learning phase of the satellite image classification. The underlying philosophy of the TSC is that an image can be more sparsely reconstructed via the images (sparse I) belonging to the same category (sparse II). The images are classified according to a newly defined “image-to-category” similarity based on the coding coefficients. Requiring no training phase, our method achieves very promising results. The experimental comparisons are shown on a real satellite image database.
208 citations
TL;DR: C-band ocean backscatter observations over operational weather buoys using RADARSAT-2 fine quad mode data have resulted in new empirical relationships for the C-band co-polarization ratio and the C.band cross-polarsization (cross-pol) ocean back scatter.
Abstract: C-band ocean backscatter observations over operational weather buoys using RADARSAT-2 fine quad mode data have resulted in new empirical relationships for the C-band co-polarization ratio and the C-band cross-polarization (cross-pol) ocean backscatter. The cross-pol relationship is independent of incidence angle and wind direction, which simplifies wind speed retrieval from synthetic aperture radar imagery for sufficiently high wind speeds.
197 citations
TL;DR: An adaptive Markov random field approach is proposed for classification of hyperspectral imagery with the introduction of a relative homogeneity index for each pixel and the use of this index to determine an appropriate weighting coefficient for the spatial contribution in the MRF classification.
Abstract: An adaptive Markov random field (MRF) approach is proposed for classification of hyperspectral imagery in this letter. The main feature of the proposed method is the introduction of a relative homogeneity index for each pixel and the use of this index to determine an appropriate weighting coefficient for the spatial contribution in the MRF classification. In this way, overcorrection of spatially high variation areas can be avoided. Support vector machines are implemented for improved class modeling and better estimate of spectral contribution to this approach. Experimental results of a synthetic hyperspectral data set and a real hyperspectral image demonstrate that the proposed method works better on both homogeneous regions and class boundaries with improved classification accuracy.
180 citations
TL;DR: In this letter, different methods to compare amplitude statistics will be presented, compared through simulation and applied to real data, and recommendations are made concerning which method to use in practice.
Abstract: Efficient estimation of the interferometric phase and complex correlation is fundamental for the full exploitation of interferometric synthetic aperture radar (InSAR) capabilities. Particularly, when combining interferometric measures arising both from distributed and concentrated targets, the interferometric phase has to be correctly extracted in order to preserve its physical meaning. Recently, an amplitude-based algorithm for the adaptive multilooking of InSAR stacks was proposed where it was shown that a comparison of the backscatter amplitude statistics is a suitable way to adaptively group and average the pixels in order to preserve the phase signatures of natural structures in the observed area. In this letter, different methods to compare amplitude statistics will be presented, compared through simulation and applied to real data. Based on these, recommendations are made concerning which method to use in practice.
179 citations
TL;DR: Simulation results show that the proposed algorithm outperforms the classical hyperspectral target detection algorithms, such as the popular spectral matched filters, matched subspace detectors, and adaptive subspace detector, as well as binary classifiers such as support vector machines.
Abstract: This letter proposes a simultaneous joint sparsity model for target detection in hyperspectral imagery (HSI). The key innovative idea here is that hyperspectral pixels within a small neighborhood in the test image can be simultaneously represented by a linear combination of a few common training samples but weighted with a different set of coefficients for each pixel. The joint sparsity model automatically incorporates the interpixel correlation within the HSI by assuming that neighboring pixels usually consist of similar materials. The sparse representations of the neighboring pixels are obtained by simultaneously decomposing the pixels over a given dictionary consisting of training samples of both the target and background classes. The recovered sparse coefficient vectors are then directly used for determining the label of the test pixels. Simulation results show that the proposed algorithm outperforms the classical hyperspectral target detection algorithms, such as the popular spectral matched filters, matched subspace detectors, and adaptive subspace detectors, as well as binary classifiers such as support vector machines.
155 citations
TL;DR: An attempt to address the main problem using the Gaussian kernel-based AD methods is the optimal setting of sigma, with a direct and adaptive measure based on a geometric interpretation of the GK-SVDD.
Abstract: Recently, anomaly detection (AD) has attracted considerable interest in a wide variety of hyperspectral remote sensing applications. The goal of this unsupervised technique of target detection is to identify the pixels with significantly different spectral signatures from the neighboring background. Kernel methods, such as kernel-based support vector data description (SVDD) (K-SVDD), have been presented as the successful approach to AD problems. The most commonly used kernel is the Gaussian kernel function. The main problem using the Gaussian kernel-based AD methods is the optimal setting of sigma. In an attempt to address this problem, this paper proposes a direct and adaptive measure for Gaussian K-SVDD (GK-SVDD). The proposed measure is based on a geometric interpretation of the GK-SVDD. Experimental results are presented on real and synthetically implanted targets of the target detection blind-test data sets. Compared to previous measures, the results demonstrate better performance, particularly for subpixel anomalies.
TL;DR: 3-D deformation reconstruction through the combination of conventional InSAR and MAI will allow for better modeling, and hence, a more comprehensive understanding, of the source geometry associated with volcanic, seismic, and other processes that are manifested by surface deformation.
Abstract: Surface deformation caused by an intrusion and small eruption during June 17-19, 2007, along the East Rift Zone of Kilauea Volcano, Hawaii, was three-dimensionally reconstructed from radar interferograms acquired by the Advanced Land Observing Satellite (ALOS) phased-array type L-band synthetic aperture radar (SAR) (PALSAR) instrument. To retrieve the 3-D surface deformation, a method that combines multiple-aperture interferometry (MAI) and conventional interferometric SAR (InSAR) techniques was applied to one ascending and one descending ALOS PALSAR interferometric pair. The maximum displacements as a result of the intrusion and eruption are about 0.8, 2, and 0.7 m in the east, north, and up components, respectively. The radar-measured 3-D surface deformation agrees with GPS data from 24 sites on the volcano, and the root-mean-square errors in the east, north, and up components of the displacement are 1.6, 3.6, and 2.1 cm, respectively. Since a horizontal deformation of more than 1 m was dominantly in the north-northwest-south-southeast direction, a significant improvement of the north-south component measurement was achieved by the inclusion of MAI measurements that can reach a standard deviation of 3.6 cm. A 3-D deformation reconstruction through the combination of conventional InSAR and MAI will allow for better modeling, and hence, a more comprehensive understanding, of the source geometry associated with volcanic, seismic, and other processes that are manifested by surface deformation.
TL;DR: Numerical simulations illustrate the performance of the algorithms, showing that the start-stop approximation may not be valid for FMCW-SAR, whereas the modified backprojection algorithm works very well here.
Abstract: In this letter, we develop time-domain reconstruction algorithms for frequency-modulated continuous wave synthetic aperture radar (FMCW-SAR). The algorithms considered here are the time-domain correlation algorithm, and two versions of the backprojection algorithm: the standard one based on the start-stop approximation, and a modified version that takes into account the movement of the sensor during the transmission of the pulse. Numerical simulations illustrate the performance of the algorithms, showing that the start-stop approximation may not be valid for FMCW-SAR, whereas the modified backprojection algorithm works very well here.
TL;DR: Results show that the combination of high-resolution data and advanced coherent processing techniques can lead to impressive reconstruction and monitoring capabilities of the whole 3-D structure of buildings.
Abstract: Layover is frequent in imaging and monitoring with synthetic aperture radar (SAR) areas characterized by a high density of scatterers with steep topography, e.g., in urban environment. Using medium-resolution SAR data tomographic techniques has been proven to be capable of separating multiple scatterers interfering (in layover) in the same pixel. With the advent of the new generation of high-resolution sensors, the layover effect on buildings becomes more evident. In this letter, we exploit the potential of the 4-D imaging applied to a set of TerraSAR-X spotlight acquisitions. Results show that the combination of high-resolution data and advanced coherent processing techniques can lead to impressive reconstruction and monitoring capabilities of the whole 3-D structure of buildings.
TL;DR: This letter proposes a strategy to deal with the thermal dilation phase component of persistent scatterer (PS) interferometry (PSI), which involves further developing the standard two-parameter PSI model with a third unknown parameter called the Thermal dilation parameter, which is estimated for each PS.
Abstract: This letter focuses on the thermal expansion component of persistent scatterer (PS) interferometry (PSI), which is a result of temperature differences in the imaged area between synthetic aperture radar (SAR) acquisitions. This letter is based on very high resolution X-band StripMap SAR data captured by the TerraSAR-X spaceborne sensor. The X-band SAR interferometric phases are highly influenced by the thermal dilation of the imaged objects. This phenomenon can have a strong impact on the PSI products, particularly on the deformation velocity maps, if not properly handled during the PSI analysis. In this letter, we propose a strategy to deal with the thermal dilation phase component, which involves further developing the standard two-parameter PSI model (deformation velocity and residual topographic error) with a third unknown parameter called the thermal dilation parameter, which is estimated for each PS. The map obtained from plotting this parameter for all PSs of a given area is hereafter called thermal map. This letter describes the proposed model and outlines the issue of parameter estimability. In addition, the potential of exploiting the thermal maps is analyzed by illustrating two examples of the Barcelona (Spain) metropolitan area. Thermal maps provide two types of information: The first one is the coefficient of thermal expansion of the observed objects, while the second one, which is related to the pattern of the thermal dilation parameter, gives information about the static structure of these objects. Two important aspects that influence the exploitation of thermal maps are discussed in the last section of this letter: the line-of-sight nature of the derived estimates and the achievable precision in the estimation of the coefficient of thermal expansion.
TL;DR: A novel unsupervised spatial preprocessing (SPP) module which adopts a region-based approach for the characterization of each endmember class prior to endmember identification using spectral information, and can be combined with any spectral-based endmember extraction technique.
Abstract: Linear spectral unmixing is an important task in remotely sensed hyperspectral data exploitation. This approach first identifies a collection of spectrally pure constituent spectra, called endmembers, and then expresses the measured spectrum of each mixed pixel as a combination of endmembers weighted by fractions or abundances that indicate the proportion of each endmember in the pixel. Over the last decade, several algorithms have been developed for automatic extraction of spectral endmembers from hyperspectral data, with many of them relying exclusively on the spectral information. In this letter, we develop a novel unsupervised spatial preprocessing (SPP) module which adopts a region-based approach for the characterization of each endmember class prior to endmember identification using spectral information. The proposed approach can be combined with any spectral-based endmember extraction technique. Our method is validated using both synthetic scenes constructed using fractals and a real hyperspectral data set collected by NASA's Airborne Visible Infrared Imaging Spectrometer over the Cuprite Mining District in Nevada and further compared with previous efforts in the same direction such as the spatial-spectral endmember extraction, automatic morphological endmember extraction, or SPP methods.
TL;DR: A novel region-location algorithm is proposed, which exploits the clustering information from matched SIFT keypoints, as well as the region information extracted through the image segmentation, which outperforms the existing algorithms in terms of detection accuracy.
Abstract: This letter presents a new method for airport detection from large high-spatial-resolution IKONOS images. To this end, we describe airport by a set of scale-invariant feature transform (SIFT) keypoints and detect it using an improved SIFT matching strategy. After obtaining SIFT matched keypoints, to both discard the redundant matched points and locate the possible regions of candidates that contain the target, a novel region-location algorithm is proposed, which exploits the clustering information from matched SIFT keypoints, as well as the region information extracted through the image segmentation. Finally, airport recognition is achieved by applying the prior knowledge to the candidate regions. Experimental results show that the proposed approach outperforms the existing algorithms in terms of detection accuracy.
TL;DR: A new method named CVA in posterior probability space (CVAPS), which analyzes the posterior probability by using CVA and shows that error cumulation in PCC was reduced and the main drawbacks in CVA were alleviated effectively by using CVAPS.
Abstract: Postclassification comparison (PCC) and change vector analysis (CVA) have been widely used for land use/cover change detection using remotely sensed data. However, PCC suffers from error cumulation stemmed from an individual image classification error, while a strict requirement of radiometric consistency in remotely sensed data is a bottleneck of CVA. This letter proposes a new method named CVA in posterior probability space (CVAPS), which analyzes the posterior probability by using CVA. The CVAPS approach was applied and validated by a case study of land cover change detection in Shunyi District, Beijing, China, based on multitemporal Landsat Thematic Mapper data. Accuracies of “change/no-change” detection and “from-to” types of change were assessed. The results show that error cumulation in PCC was reduced in CVAPS. Furthermore, the main drawbacks in CVA were also alleviated effectively by using CVAPS. Therefore, CVAPS is potentially useful in land use/cover change detection.
TL;DR: A time-efficient solution to estimate the error of satellite surface soil moisture from the land parameter retrieval model is presented and could substitute computationally intensive methods for near-real-time data assimilation studies where both the soil moisture product and error estimate are needed.
Abstract: A time-efficient solution to estimate the error of satellite surface soil moisture from the land parameter retrieval model is presented. The errors are estimated using an analytical solution for soil moisture retrievals from this radiative-transfer-based model that derives soil moisture from low-frequency passive microwave observations. The error estimate is based on a basic error propagation equation which uses the partial derivatives of the radiative transfer equation and estimated errors for each individual input parameter. Results similar to those of the Monte Carlo approach show that the developed time-efficient methodology could substitute computationally intensive methods. This procedure is therefore a welcome solution for near-real-time data assimilation studies where both the soil moisture product and error estimate are needed. The developed method is applied to the C-, X-, and Ku-bands of the Aqua/Advanced Microwave Scanning Radiometer for Earth Observing System sensor to study differences in errors between frequencies.
TL;DR: A novel algorithm of inverse synthetic aperture radar (ISAR) imaging based on Bayesian estimation is proposed, wherein the ISAR imaging joint with phase adjustment is mathematically transferred into signal reconstruction via maximum a posteriori estimation.
Abstract: In this letter, a novel algorithm of inverse synthetic aperture radar (ISAR) imaging based on Bayesian estimation is proposed, wherein the ISAR imaging joint with phase adjustment is mathematically transferred into signal reconstruction via maximum a posteriori estimation. In the scheme, phase errors are treated as model errors and are overcome in the sparsity-driven optimization regardless of the formats, while data-driven estimation of the statistical parameters for both noise and target is developed, which guarantees the high precision of image generation. Meanwhile, the fast Fourier transform is utilized to implement the solution to image formation, promoting its efficiency effectively. Due to the high denoising capability of the proposed algorithm, high-quality image also could be achieved even under strong noise. The experimental results using simulated and measured data confirm the validity.
TL;DR: A modified KDA algorithm, i.e., kernel local Fisher discriminant analysis (KLFDA), is studied for HSI analysis and classification experiments demonstrate that this approach outperforms current state-of-the-art HSI-classification methods.
Abstract: Linear discriminant analysis (LDA) has been widely applied for hyperspectral image (HSI) analysis as a popular method for feature extraction and dimensionality reduction. Linear methods such as LDA work well for unimodal Gaussian class-conditional distributions. However, when data samples between classes are nonlinearly separated in the input space, linear methods such as LDA are expected to fail. The kernel discriminant analysis (KDA) attempts to address this issue by mapping data in the input space onto a subspace such that Fisher's ratio in an intermediate (higher-dimensional) kernel-induced space is maximized. In recent studies with HSI data, KDA has been shown to outperform LDA, particularly when the data distributions are non-Gaussian and multimodal, such as when pixels represent target classes severely mixed with background classes. In this letter, a modified KDA algorithm, i.e., kernel local Fisher discriminant analysis (KLFDA), is studied for HSI analysis. Unlike KDA, KLFDA imposes an additional constraint on the mapping-it ensures that neighboring points in the input space stay close-by in the projected subspace and vice versa. Classification experiments with a challenging HSI task demonstrate that this approach outperforms current state-of-the-art HSI-classification methods.
TL;DR: The multicomponent generalization of time warp rewrites the D-TomoSAR system model to an (M + 1)-dimensional standard spectral estimation problem, where M indicates the user-defined motion model order and enables the motion estimation for all possible complex motion models.
Abstract: In the differential synthetic aperture radar tomography (D-TomoSAR) system model, the motion history appears as a phase term. In the case of nonlinear motion, this phase term is no longer linear and, hence, cannot be retrieved by spectral estimation. We propose the “time warp” method that rearranges the acquisition dates such that a linear motion is pretended. The multicomponent generalization of time warp rewrites the D-TomoSAR system model to an (M + 1)-dimensional standard spectral estimation problem, where M indicates the user-defined motion model order and, hence, enables the motion estimation for all possible complex motion models. Both simulations and real data (from TerraSAR-X spotlight) examples demonstrate the applicability of the method and show that linear and periodic (seasonal) motion components can be separated and retrieved.
TL;DR: Preliminary results of an attempt to monitor rice crop growth using RADARSAT-2 quad polarization synthetic aperture radar (SAR) data show that an HV or VH image backscattering coefficient exhibits the best correlation with rice age after transplantation.
Abstract: This letter presents preliminary results of an attempt to monitor rice crop growth using RADARSAT-2 quad polarization synthetic aperture radar (SAR) data. Three RADARSAT-2 quad-polarization SAR images are collected from transplanting to rice crop harvesting. Ground truth data, such as rice height and biomass, are measured during RADARSAT-2 data acquisition in Hainan Province, South China. The correlation between backscattering coefficient and rice growth parameters is analyzed, and then, a rice field mapping method with quad polarization SAR image is developed. Experiments show that an HV or VH image backscattering coefficient exhibits the best correlation with rice age after transplantation. Furthermore, the HV or VH image is also more suitable for retrieving rice growth parameters, such as rice height and dry biomass, for FQ4 RADARSAT-2 SAR data. The ratio image of HH/VV possesses high separability required to distinguish rice crop from banana, forest, and river. Results indicate that RADARSAT-2 quad polarization SAR data presented enormous potential for monitoring rice crop growth.
TL;DR: A new ISAR imaging algorithm based on the product generalized cubic phase function (PGCPF) is proposed, which can estimate the parameters of a multicomponent cubic phase signal, and combined with the range-instantaneous-Doppler technique, high-quality instantaneous ISAR images can be obtained.
Abstract: For inverse synthetic aperture radar (ISAR) imaging of a maneuvering target, the Doppler frequency shift for the received signal in a range bin is time varying. In this letter, the received signal is modeled as a multicomponent cubic phase signal, and a new ISAR imaging algorithm based on the product generalized cubic phase function (PGCPF) is proposed. The PGCPF algorithm can estimate the parameters of a multicomponent cubic phase signal, and combined with the range-instantaneous-Doppler technique, high-quality instantaneous ISAR images can be obtained. Results of real data demonstrate the effectiveness of the new method presented in this letter.
TL;DR: A model-based analysis demonstrates that the LiDAR waveforms cannot only capture the tree height information but also picks up the seasonal and vertical variation of NDVI inside the tree canopy.
Abstract: The first demonstration of a multispectral light detection and ranging (LiDAR) optimized for detailed structure and physiology measurements in forest ecosystems is described. The basic principle is to utilize, in a single instrument, both the capacity of multispectral sensing to measure plant physiology [through normalized difference vegetation index (NDVI) and photochemical reflectance index (PRI)] with the ability of LiDAR to measure vertical structure information and generate “hot spot” (specular) reflectance data independent of solar illumination. A tunable laser operated at four wavelengths (531, 550, 660, and 780 nm) was used to measure profiles of the NDVI and the PRI. Laboratory-based measurements were conducted for live trees, demonstrating that realistic values of the indexes can be measured. A model-based analysis demonstrates that the LiDAR waveforms cannot only capture the tree height information but also picks up the seasonal and vertical variation of NDVI inside the tree canopy.
TL;DR: This letter proposes a Parzen-window-kernel-based algorithm for ship detection in synthetic aperture radar (SAR) images that uses the data-driving kernel functions of Parzen window to approximate the histogram of real SAR image.
Abstract: This letter proposes a Parzen-window-kernel-based algorithm for ship detection in synthetic aperture radar (SAR) images. First, the data-driving kernel functions of Parzen window are utilized to approximate the histogram of real SAR image, in order to complete the accurate modeling of SAR images. Then, a threshold of global constant false alarm rate is given theoretically, and the numerical solution of the threshold is also derived. The experimental results of the real data of typical targets demonstrate that the algorithm presented is effective.
TL;DR: The experimental results show that the proposed semisupervised band clustering algorithm can outperform other existing methods with lower computational cost.
Abstract: Band clustering is applied to dimensionality reduction of hyperspectral imagery. Different from unsupervised clustering using all the pixels or supervised clustering requiring labeled pixels, the proposed semisupervised band clustering needs class spectral signatures only. After clustering, a cluster selection step is applied to select clusters to be used in the following data analysis. Initial conditions and distance metrics are also investigated to improve the clustering performance. The experimental results show that the proposed algorithm can outperform other existing methods with lower computational cost.
TL;DR: An experimental analysis of the application of the ε-insensitive support vector regression (SVR) technique to soil moisture content estimation from remotely sensed data at field/basin scale provides useful indications for building soil moisture estimation processors for upcoming satellites or near-real-time applications.
Abstract: This letter presents an experimental analysis of the application of the e-insensitive support vector regression (SVR) technique to soil moisture content estimation from remotely sensed data at field/basin scale. SVR has attractive properties, such as ease of use, good intrinsic generalization capability, and robustness to noise in the training data, which make it a valid candidate as an alternative to more traditional neural-network-based techniques usually adopted in soil moisture content estimation. Its effectiveness in this application is assessed by using field measurements and considering various combinations of the input features (i.e., different active and/or passive microwave measurements acquired using various sensor frequencies, polarizations, and acquisition geometries). The performance of the SVR method (in terms of estimation accuracy, generalization capability, computational complexity, and ease of use) is compared with that achieved using a multilayer perceptron neural network, which is considered as a benchmark in the addressed application. This analysis provides useful indications for building soil moisture estimation processors for upcoming satellites or near-real-time applications.
TL;DR: It is shown that effective roughness parameters are a promising tool for soil moisture retrieval under a wheat canopy and that the use of a leaf area index may be recommended above other vegetation indicators, as it leads to the lowest root-mean-square errors of about 5.5 vol%.
Abstract: The synthetic aperture radar (SAR)-based soil moisture retrieval of agricultural fields is often hampered by vegetation effects on the backscattered signal. The semiempirical water cloud model (WCM) allows for estimating the backscatter of a vegetated surface, accounting for both the contributions of the vegetation and the underlying soil. The latter is often described through the integral equation model (IEM). Unfortunately, the IEM requires an accurate parameterization of the surface roughness which is very difficult to achieve. Therefore, this letter extends the WCM with a bare soil contribution that is based on the IEM, which, however, relies on calibrated or effective roughness parameters. Furthermore, this letter compares a number of vegetation indicators for their use in the WCM. Based on a series of L-band SAR observations, it is shown that effective roughness parameters are a promising tool for soil moisture retrieval under a wheat canopy and that the use of a leaf area index may be recommended above other vegetation indicators, as it leads to the lowest root-mean-square errors of about 5.5 vol%. These results prove the operational potential of L-band SAR data for soil moisture inferred under a wheat canopy throughout the entire crop growth cycle.
TL;DR: An automated scheme for eddy detection from remote sensing SST data is presented based on the analysis of velocity fields derived from SST measurements (thermal-wind velocity field), which can identify positions of eddy centers and derive eddy size, intensity, path, and lifetime.
Abstract: Cyclonic (anticyclonic) oceanic eddies drive local upwelling (downwelling), leaving footprints in the sea surface temperature (SST) field as local extremes. Satellite-measured SST images can therefore be used to obtain information of the characteristics of oceanic eddies. Remotely sensed measurements represent very large data sets, both spatially and temporally. Manual eddy detection and analysis are thus practically impossible. In this letter, an automated scheme for eddy detection from remote sensing SST data is presented. The method is based on the analysis of velocity fields derived from SST measurements (thermal-wind velocity field). Using the geometric features of the velocity field, we can identify positions of eddy centers and derive eddy size, intensity, path, and lifetime. The scheme is applied to a realistic remotely sensed SST data set in a strong eddy activity region: Kuroshio Extension region.