Hyperspectral unmixing aims at determining the reference spectral signatures composing a hyperspectral image, their abundance fractions and their number. In practice, the spectral variability of the identified signatures induces significant abundance estimation errors. To address this issue, this paper introduces a new linear mixing model explicitly accounting for this phenomenon. In this setting, the extracted endmembers are interpreted as possibly corrupted versions of the true endmembers. The parameters of this model can be estimated using an optimization algorithm based on the alternating direction method of multipliers. The performance of the proposed unmixing method is evaluated on synthetic and real data.

Texture analysis is central in many image processing problems. It can be conducted by studying the local regularity fluctuations of image amplitudes, and multifractal analysis provides a theoretical and practical framework for such a characterization. Yet, due to the non Gaussian nature and intricate dependence structure of multifractal models, accurate parameter estimation is challenging : standard estimators yield modest performance, and alternative (semi-)parametric estimators exhibit prohibitive computational cost for large images. This present contribution addresses these difficulties and proposes a Bayesian procedure for the estimation of the multifractality parameter c2 for images. It relies on a recently proposed semi-parametric model for the multivariate statistics of log-wavelet leaders and on a Whittle approximation that enables its numerical evaluation. The key result is a closed-form expression for the Whittle likelihood. Numerical simulations indicate the excellent performance of the method, significantly improving estimation performance over standard estimators and computational efficiency over previously proposed Bayesian estimators.

The profusion of spectral bands generated by the acquisition process of hyperspectral images generally leads to high computational costs. Such difficulties arise in particular with nonlinear unmixing methods, which are naturally more complex than linear ones. This complexity, associated with the high redundancy of information within the complete set of bands, make the search of band selection algorithms relevant. With this work, we propose a band selection strategy in reproducing kernel Hilbert spaces that allows to drastically reduce the processing time required by nonlinear unmixing techniques. Simulation results show a complexity reduction of two orders of magnitude without compromising unmixing performance.

Electrical engineering teaching is not an easy task because of the broad spectrum of knowledge to call for (electromagnetic, electronic, control, signal processing), each one having its specific formalism. To connect these different courses through a real-life application, we have decided to design a fire-control model based on a low-cost sonar system. This experiment has been designed for graduated students and is exploited in laboratory projects. Besides the playful aspects brought by the model, the project allows to face-off a real system and requires strong initiative from the students to success.

The specifications of Digital Video Broadcasting - Return Channel via Satellite (DVB-RCS2) state that the satellite gateway could introduce both random and dedicated access methods to distribute the capacity among the different home users. Before starting an engineering process to design an algorithm allowing to combine both methods, it seems necessary to assess the performance of each. This paper compares random and dedicated access methods by measuring their impact on the performance of Transmission Control Protocol (TCP) sessions when the home users exploit the DVB-RCS2 link for regular use (e.g., web browsing or email transmission). In this paper we detail the implementation of an NS-2 module emulating Physical Channel Access (PCA). This module fills a gap in terms of random and deterministic access methods and allows to model various satellite channel access strategies. Based on NS-2 simulations using realistic system parameters of the DVB-RCS2 link, we demonstrate that, compared to dedicated access methods, which generally result in higher levels of transmitted data, random access methods enable faster transmission for short flows. We propose to combine random and dedicated access methods, with the selection of a specific method dependent on the dynamic load of the network and the sequence number of the TCP segments.

This paper presents a Bayesian fusion technique for remotely sensed multi-band images. The observed images are related to the high spectral and high spatial resolution image to be recovered through physical degradations, e.g., spatial and spectral blurring and/or subsampling defined by the sensor characteristics. The fusion problem is formulated within a Bayesian estimation framework. An appropriate prior distribution exploiting geometrical considerations is introduced. To compute the Bayesian estimator of the scene of interest from its posterior distribution, a Markov chain Monte Carlo algorithm is designed to generate samples asymptotically distributed according to the target distribution. To efficiently sample from this high-dimension distribution, a Hamiltonian Monte Carlo step is introduced within a Gibbs sampling strategy. The efficiency of the proposed fusion method is evaluated with respect to several state-of-the-art fusion techniques.

AStrion is an automatic spectrum analyser software, which proposes a new generic and data-driven method without any a priori information on the measured signals. In order to compute some general characteristics and derive the properties of the signal, the first step in this approach is to give some insight into the nature of the signal. This pre-analysis, the so-called data validation, contains a number of tests to reveal some of the properties and characteristics of the data, such as the acquisition validity (the absence of saturation and a posteriori in respect of the sampling theorem), the stationarity (or non-stationarity), the periodicity and the signal-to-noise ratio. Based on these characteristics, the proposed method defines indicators and alarm trigger thresholds and also categorises the signal into three classes, which helps to guide the following spectral analysis. The present paper introduces the four tests of this pre-analysis and the signal categorisation rules. Finally, the proposed approach is validated on a set of wind turbine vibration measurements to demonstrate its applicability for a long-term and continuous monitoring of real-world signals.

This paper proposes an automatic procedure for condition monitoring. It represents a valuable tool for maintenance of expensive and spread systems such as wind turbine farms. Thanks to data-driven signal processing algorithms, the proposed solution is fully automatic for the user. The paper briefly describes all the steps of the processing, from pre-processing of acquired signal to interpretation of generated results. It starts with an angular resampling method with speed measurement correction. Then comes a data validation step, in both time/angular and frequency/order domains. After these preprocessings, the spectral components of the analyzed signal are identified and classified in several classes from sine wave to narrow band components. This spectral peak detection and classification allows extracting the harmonic and side-band series which may be part of the signal spectral content. Moreover, the detected spectral patterns are associated with the characteristic frequencies of the investigated system. Based on the detected side-band series, the full-band demodulation is performed. At each step, the diagnosis features are computed and dynamically tracked signal by signal. Finally, system health indicators are proposed to conclude about the condition of the investigated system. All mentioned steps create a self-sufficient tool for robust diagnosis of mechanical faults. The paper presents the performance of the proposed method on real-world signals from a wind turbine drive train.

Texture characterization is a central element in many image processing applications. Multifractal analysis is a useful signal and image processing tool, yet, the accurate estimation of multifractal parameters for image texture remains a challenge. This is due in the main to the fact that current estimation procedures consist of performing linear regressions across frequency scales of the 2D dyadic wavelet transform, for which only a few such scales are computable for images. The strongly non-Gaussian nature of multifractal processes, combined with their complicated dependence structure, makes it difficult to develop suitable models for parameter estimation. Here, we propose a Bayesian procedure that addresses the difficulties in the estimation of the multifractality parameter. The originality of the procedure is threefold. The construction of a generic semiparametric statistical model for the logarithm of wavelet leaders; the formulation of Bayesian estimators that are associated with this model and the set of parameter values admitted by multifractal theory; the exploitation of a suitable Whittle approximation within the Bayesian model which enables the otherwise infeasible evaluation of the posterior distribution associated with the model. Performance is assessed numerically for several 2D multifractal processes, for several image sizes and a large range of process parameters. The procedure yields significant benefits over current benchmark estimators in terms of estimation performance and ability to discriminate between the two most commonly used classes of multifractal process models. The gains in performance are particularly pronounced for small image sizes, notably enabling for the first time the analysis of image patches as small as 64 × 64 pixels.

Pansharpening first aims at fusing a panchromatic image with a multispectral image to generate an image with the high spatial resolution of the former and the spectral resolution of the latter. In the last decade many algorithms have been presented in the literature for pansharpening using multispectral data. With the increasing availability of hyperspectral systems these methods are now extending to hyperspectral images. In this work, we attempt to compare new pansharpening techniques designed for hyperspectral data with some of the state of the art methods for multispectral pansharpening, which have been adapted for hyperspectral data. Nine methods from different classes are analysed: component substitution, multiresolution analysis, hybrid, Bayesian and matrix decomposition approaches. These techniques are evaluated with the Wald’s Procol on one dataset to characterize their performances and their robustness.