No abstract

Remotely sensed hyperspectral image classification is a very challenging task due to the spatial correlation of the spectral signature and the high cost of true sample labeling. In light of this, the collective inference paradigm allows us to manage the spatial correlation between spectral responses of neighboring pixels, as interacting pixels are labeled simultaneously. The transductive inference paradigm allows us to reduce the inference error for the given set of unlabeled data, as sparsely labeled pixels are learned by accounting for both labeled and unlabeled information. In this paper, both these paradigms contribute to the definition of a spectral-relational classification methodology for imagery data. We propose a novel algorithm to assign a class to each pixel of a sparsely labeled hyperspectral image. It integrates the spectral information and the spatial correlation through an ensemble system. For every pixel of a hyperspectral image, spatial neighborhoods are constructed and used to build application-specific relational features. Classification is performed with an ensemble comprising a classifier learned by considering the available spectral information (associated with the pixel) and the classifiers learned by considering the extracted spatio-relational information (associated with the spatial neighborhoods). The more reliable labels predicted by the ensemble are fed back to the labeled part of the image. Experimental results highlight the importance of the spectral-relational strategy for the accurate transductive classification of hyperspectral images and they validate the proposed algorithm.

Fully-focusing of radar altimeters is a recent concept that has been introduced to allow further improvement of along-track resolution in high pulse repetition frequency (PRF) radar altimeters. The straight potentiality of this new perspective reflects into a more accurate estimation of geophysical parameters in some applications such as sea-ice observation. However, as documented in a recent paper, such capability leaves unsolved the problem of the high computational effort required. In this paper, we face the problem of adapting for altimeters the Omega-Kappa SAR focusing algorithm that is performed in the two-dimensional wavenumber domain, accounting for the difference existing between SAR and altimeter from geometry (looking and swath width) and instrument (echoes are deramped onboard on receiving) point of view. Simulations and an application using in-orbit data show the effectiveness of the proposed approach and the highly reduced computational effort.

Hydrogel composite membranes (HCMs) are used as novel mineralization platforms for the bioinspired synthesis of CaCO3 superstructures. A comprehensive statistical analysis of the experimental results reveals quantitative relationships between crystallization conditions and crystal texture and a strong selectivity toward complex morphologies when monomers bearing carboxyl and hydroxyl groups are used together in the hydrogel layer synthesis in HCMs.

The letter introduces a novel quantizer suited for medium to high-resolution synthetic aperture radar (SAR) systems, like the forthcoming SENTINEL-1 SAR. The Flexible Dynamic Block Adaptive Quantization (FDBAQ) extends the concept of the Block Adaptive Quantization (BAQ), used in spaceborne SAR since the Magellan mission, by adaptively tuning the quantizer rate according to the local signal-to-noise-ratio (SNR). A design is presented aiming to optimize the average bit-rate, while constraining the minimum SNR. FDBAQ optimized performance is then evaluated using backscatter maps derived from ENVIronment SATellite (ENVISAT) data

The paper analyses an along-track multistatic Synthetic Aperture Radar (SAR) formation. The formation aims at achieving a high azimuth resolution maintaining at the same time a large swath width. The case with one transmitting sensor and all receiving is analyzed (Single Input Multiple Output, SIMO). An effective and novel reconstruction, in the two-dimensional frequency domain is introduced that is able to keep low the azimuth ambiguity and achieve a recombination gain close to the theoretical one. Degradation of the system performance due to the loss of the control of formation position is analyzed using probabilistic considerations. Moreover, some innovative methods to mitigate the loss of optimality are introduced and evaluated using simulations. Finally, considerations on the impact of the across-track non zero baseline are discussed.

Solid-state reactivity is often studied by in situ experiments with a multi-technique approach, where complementarity of different probes is exploited. In situ data are usually analysed using a complex protocol: first the reaction model most suited to describe the specific solid-state reaction is chosen, second the reaction coordinate is obtained from the data, the order of reaction is then calculated by applying a specific kinetic equation, and finally kinetic parameters are obtained with an Arrhenius plot. The approach is both time consuming and subject to errors due to the arbitrariness of extraction of the reaction coordinate, typically from individual peak intensity variations during the reaction. In addition, application of the different kinetic equations to obtain the best fitting one is tedious and no general method to select the best model with an unbiased approach is available. Here we propose a new procedure based on principal component analysis to get kinetic information from in situ data, which simplifies and speeds up the process of kinetic parameter calculation from a three- to a two- or even a one-step form, reaching a high degree of automation and the ability to manage the huge amount of data produced by in situ multi-technique experiments. The new approach treats data as a whole, without biases introduced by manual methods of obtaining the reaction coordinate by peak intensity evaluation from individual patterns typical of the traditional approach. The procedure is described in its theoretical framework and applied to the formation of a molecular complex, monitored by in situ X-ray powder diffraction and Raman measurements.

The development of two solid-state reactions, Xe absorption into MFI and molecular complex formation, where samples are affected by changes of crystal lattice due to temperature or pressure variation was structurally monitored through in situ or in operando X-ray powder diffraction experiments. Consequent variations of the peak positions prevent collective analysis of measured patterns, aiming at investigating structural changes occurring within the crystal cell. Moreover, an intrinsic and variable error in peak position is unavoidable when using the Bragg-Brentano geometry and, in some cases (sticky, bulky, aggregate samples) the sample mounting can increase the error within a dataset. Here we present a general multivariate analysis method to process in a fast and automatic way in situ XRPD data collected on charge transfer complexes and porous materials, with the capacity of disentangling peak shifts from intensity and shape variations in diffraction signals, thus allowing an efficient separation of the contribution of crystal lattice changes from structural changes. The peak shift correction allowed an improved PCA analysis that turned out to be more sensible than the traditional single pattern Rietveld analysis. The developed algorithms allowed, with respect to the traditional approach, the location of two new Xe positions into MFI with a better interpretation of the experimental data, while a much faster and more efficient recovery of the reaction coordinate was achieved in the molecular complex formation reaction.