Search

The imbalance between the (often limited) number of available training samples and the high data dimensionality, together with the presence of mixed pixels, often complicates the classification of remotely sensed hyperspectral data. In this paper, we tackle these problems by developing a new method that combines spectral unmixing and classification techniques in a subspace-based approach. The proposed method is developed under the assumption that the spectral signature of a land cover class is associated with a given set of pure spectral signatures (called endmembers in spectral unmixing terminology), which define a low-dimensional subspace with clear physical meaning. We aim to exploit this relationship to learn the class-dependent subspaces and integrate them with a multinomial logistic regression procedure. Experiments on synthetic datasets and real hyperspectral images show that our method is able to obtain competitive performances in comparison with other approaches, particularly when very limited training sets are available. [ABSTRACT FROM AUTHOR]

Accurately estimating the elements in Earth observations is crucial when assessing specific features such as air quality index, water pollution, or urbanization process behavior. Moreover, physical–chemical composition can be retrieved from hyperspectral images when proper spectral unmixing architectures are employed. Specifically, when linear and nonlinear combinations of endmembers (pure spectral components) are accurately characterized, hyperspectral unmixing plays a key role in understanding and quantifying phenomena occurring over the instantaneous field-of-view. Thus, reliable detection of nonlinear reflectance behavior can play a key role in enhancing hyperspectral unmixing performance. In this paper, two new methods for adaptive design of mixture models for hyperspectral unmixing are introduced. One of the methods relies on exploiting geometrical features of hyperspectral signatures in terms of nonorthogonal projections onto the space induced by the endmembers’ spectra. Then, an iterative process aims at understanding the order of local nonlinearity that is displayed by each endmember over every pixel. An improved version of an artificial neural network-based approach for nonlinearity order information is also considered and compared. Experimental results show that the proposed approaches are actually able to retrieve thorough information on the nature of the nonlinear effects over the image, while providing excellent performance in reconstructing the given data sets. [ABSTRACT FROM AUTHOR]

In this letter, a new method for higher order nonlinear hyperspectral unmixing is introduced. The proposed scheme relies on the harmonic description of the endmembers contributions to characterize the interactions among the materials showing up in the given scenes. Moreover, it aims at directly estimating the probability of occurrence of each material in the images, so to provide an accurate quantification of the endmembers also in complex scenarios. Experimental results carried out on synthetic and real data sets show that the proposed method is able to obtain good unmixing performance when compared to other state-of-the-art architectures. [ABSTRACT FROM AUTHOR]

To efficiently model high-order nonlinear material mixtures in complex scenery, more and more complex spectral mixing models have been developed, so that over-fitting phenomena more often occur during the unmixing process. Therefore, the accurate and robust inversion of material abundances is a challenging task, especially for low signal-to-noise ratio (SNR) data. In this paper, this task is achieved by inverting the parameters using a hierarchical Bayesian model based on the P-linear mixing model (PLMM). Moreover, spatial information is integrated in the inversion process by considering that similar pixels share the same prior information. Thanks to the fact that PLMM can be translated into a linear model using endmembers and their powers, unmixing is performed by solving a convex optimization problem. Results obtained from synthetic and real data show that the proposed algorithm improves the accuracy of abundance estimation and efficiently reduces over-fitting effects in low SNR data. [ABSTRACT FROM PUBLISHER]

Algorithms allowing the deconvolution of hyperspectral data play a key-role in remotely sensed data processing for mineralogical investigation. Modified Gaussian model (MGM) based methods are of particular interest because they are able to retrieve accurate estimates of minerals abundances and chemistry in surface's rocks. However, MGM-based frameworks deliver high computational complexity and sensitivity to initial parameters for statistical distribution definition. In this paper, a new approach for efficient and robust mineralogical investigation over extraterrestrial bodies is introduced. The proposed framework takes advantage of the solid characterization of remote sensing hyperspectral images by unmixing higher order nonlinear combinations of reflectance features associated with mafic minerals. Experimental results achieved over Mars and Moon hyperspectral images show that the proposed scheme is able to retrieve magmatic mineral abundance maps that are highly correlated to those achieved by means of MGM-based scheme while overcoming the aforesaid issues. Finally, an empirical study allowing to distinguish between clino- and orthopyroxenes by properly processing the outcomes of nonlinear hyperspectral unmixing method is reported. [ABSTRACT FROM PUBLISHER]

In order to achieve a better knowledge of the effect of the anthropogenic extents over the environment, extracting reliable and effective information by Earth observations (EOs) is crucial to help developing a sound human–environment interaction (HEI) assessment. In this sense, the use of future hyperspectral sensors for wide area characterization leads to the need of hyperspectral unmixing (HSU) architectures to recognize urban materials and structures. Further, as urban settlements are often characterized by geometrically and spectrally complex scenarios, the nonlinear reflectance interplay among the elements that constitute each scene must be very well detailed and described so that a thorough knowledge of the scenes can be carried out. In this paper, properly set higher order nonlinear mixture models are used to perform an accurate characterization of the anthropogenic settlements in several EO scenes acquired in different continents. Moreover, a brand new index for estimation of urban extents is provided. Experimental results show how the proposed approach is able to deliver accurate and reliable characterization of urban materials and extents. [ABSTRACT FROM PUBLISHER]