Unmixing hyperspectral data: independent and dependent component analysis

The development of high spatial resolution airborne and spaceborne sensors has improved the capability of ground-based data collection in the fields of agriculture, geography, geology, mineral identification, detection [2, 3], and classification [4–8]. The signal read by the sensor from a given spatial element of resolution and at a given spectral band is a mixing of components originated by the constituent substances, termed endmembers, located at that element of resolution. This chapter addresses hyperspectral unmixing, which is the decomposition of the pixel spectra into a collection of constituent spectra, or spectral signatures, and their corresponding fractional abundances indicating the proportion of each endmember present in the pixel [9, 10]. Depending on the mixing scales at each pixel, the observed mixture is either linear or nonlinear [11, 12]. The linear mixing model holds when the mixing scale is macroscopic [13]. The nonlinear model holds when the mixing scale is microscopic (i.e., intimate mixtures) [14, 15]. The linear model assumes negligible interaction among distinct endmembers [16, 17]. The nonlinear model assumes that incident solar radiation is scattered by the scene through multiple bounces involving several endmembers [18]. Under the linear mixing model and assuming that the number of endmembers and their spectral signatures are known, hyperspectral unmixing is a linear problem, which can be addressed, for example, under the maximum likelihood setup [19], the constrained least-squares approach [20], the spectral signature matching [21], the spectral angle mapper [22], and the subspace projection methods [20, 23, 24]. Orthogonal subspace projection [23] reduces the data dimensionality, suppresses undesired spectral signatures, and detects the presence of a spectral signature of interest. The basic concept is to project each pixel onto a subspace that is orthogonal to the undesired signatures. As shown in Settle [19], the orthogonal subspace projection technique is equivalent to the maximum likelihood estimator. This projection technique was extended by three unconstrained least-squares approaches [24] (signature space orthogonal projection, oblique subspace projection, target signature space orthogonal projection). Other works using maximum a posteriori probability (MAP) framework [25] and projection pursuit [26, 27] have also been applied to hyperspectral data. In most cases the number of endmembers and their signatures are not known. Independent component analysis (ICA) is an unsupervised source separation process that has been applied with success to blind source separation, to feature extraction, and to unsupervised recognition [28, 29]. ICA consists in finding a linear decomposition of observed data yielding statistically independent components. Given that hyperspectral data are, in given circumstances, linear mixtures, ICA comes to mind as a possible tool to unmix this class of data. In fact, the application of ICA to hyperspectral data has been proposed in reference 30, where endmember signatures are treated as sources and the mixing matrix is composed by the abundance fractions, and in references 9, 25, and 31–38, where sources are the abundance fractions of each endmember. In the first approach, we face two problems: (1) The number of samples are limited to the number of channels and (2) the process of pixel selection, playing the role of mixed sources, is not straightforward. In the second approach, ICA is based on the assumption of mutually independent sources, which is not the case of hyperspectral data, since the sum of the abundance fractions is constant, implying dependence among abundances. This dependence compromises ICA applicability to hyperspectral images. In addition, hyperspectral data are immersed in noise, which degrades the ICA performance. IFA [39] was introduced as a method for recovering independent hidden sources from their observed noisy mixtures. IFA implements two steps. First, source densities and noise covariance are estimated from the observed data by maximum likelihood. Second, sources are reconstructed by an optimal nonlinear estimator. Although IFA is a well-suited technique to unmix independent sources under noisy observations, the dependence among abundance fractions in hyperspectral imagery compromises, as in the ICA case, the IFA performance. Considering the linear mixing model, hyperspectral observations are in a simplex whose vertices correspond to the endmembers. Several approaches [40–43] have exploited this geometric feature of hyperspectral mixtures [42]. Minimum volume transform (MVT) algorithm [43] determines the simplex of minimum volume containing the data. The MVT-type approaches are complex from the computational point of view. Usually, these algorithms first find the convex hull defined by the observed data and then fit a minimum volume simplex to it. Aiming at a lower computational complexity, some algorithms such as the vertex component analysis (VCA) [44], the pixel purity index (PPI) [42], and the N-FINDR [45] still find the minimum volume simplex containing the data cloud, but they assume the presence in the data of at least one pure pixel of each endmember. This is a strong requisite that may not hold in some data sets. In any case, these algorithms find the set of most pure pixels in the data. Hyperspectral sensors collects spatial images over many narrow contiguous bands, yielding large amounts of data. For this reason, very often, the processing of hyperspectral data, included unmixing, is preceded by a dimensionality reduction step to reduce computational complexity and to improve the signal-to-noise ratio (SNR). Principal component analysis (PCA) [46], maximum noise fraction (MNF) [47], and singular value decomposition (SVD) [48] are three well-known projection techniques widely used in remote sensing in general and in unmixing in particular. The newly introduced method [49] exploits the structure of hyperspectral mixtures, namely the fact that spectral vectors are nonnegative. The computational complexity associated with these techniques is an obstacle to real-time implementations. To overcome this problem, band selection [50] and non-statistical [51] algorithms have been introduced. This chapter addresses hyperspectral data source dependence and its impact on ICA and IFA performances. The study consider simulated and real data and is based on mutual information minimization. Hyperspectral observations are described by a generative model. This model takes into account the degradation mechanisms normally found in hyperspectral applications—namely, signature variability [52–54], abundance constraints, topography modulation, and system noise. The computation of mutual information is based on fitting mixtures of Gaussians (MOG) to data. The MOG parameters (number of components, means, covariances, and weights) are inferred using the minimum description length (MDL) based algorithm [55]. We study the behavior of the mutual information as a function of the unmixing matrix. The conclusion is that the unmixing matrix minimizing the mutual information might be very far from the true one. Nevertheless, some abundance fractions might be well separated, mainly in the presence of strong signature variability, a large number of endmembers, and high SNR. We end this chapter by sketching a new methodology to blindly unmix hyperspectral data, where abundance fractions are modeled as a mixture of Dirichlet sources. This model enforces positivity and constant sum sources (full additivity) constraints. The mixing matrix is inferred by an expectation-maximization (EM)-type algorithm. This approach is in the vein of references 39 and 56, replacing independent sources represented by MOG with mixture of Dirichlet sources. Compared with the geometric-based approaches, the advantage of this model is that there is no need to have pure pixels in the observations. The chapter is organized as follows. Section 6.2 presents a spectral radiance model and formulates the spectral unmixing as a linear problem accounting for abundance constraints, signature variability, topography modulation, and system noise. Section 6.3 presents a brief resume of ICA and IFA algorithms. Section 6.4 illustrates the performance of IFA and of some well-known ICA algorithms with experimental data. Section 6.5 studies the ICA and IFA limitations in unmixing hyperspectral data. Section 6.6 presents results of ICA based on real data. Section 6.7 describes the new blind unmixing scheme and some illustrative examples. Section 6.8 concludes with some remarks.

Hyperspectral unmixing algorithm via dependent component analysis

This paper introduces a new method to blindly unmix hyperspectral data, termed dependent component analysis (DECA). This method decomposes a hyperspectral images into a collection of reflectance (or radiance) spectra of the materials present in the scene (endmember signatures) and the corresponding abundance fractions at each pixel. DECA assumes that each pixel is a linear mixture of the endmembers signatures weighted by the correspondent abundance fractions. These abudances are modeled as mixtures of Dirichlet densities, thus enforcing the constraints on abundance fractions imposed by the acquisition process, namely non-negativity and constant sum. The mixing matrix is inferred by a generalized expectation-maximization (GEM) type algorithm. This method overcomes the limitations of unmixing methods based on Independent Component Analysis (ICA) and on geometrical based approaches. The effectiveness of the proposed method is illustrated using simulated data based on U.S.G.S. laboratory spectra and real hyperspectral data collected by the AVIRIS sensor over Cuprite, Nevada.

A lightweight distributed super peer election algorithm for unstructured dynamic P2P systems

Dissertação apresentada na Faculdade de Ciências e Tecnologia da Universidade Nova de Lisboa para obtenção do grau de Mestre em Engenharia Electrotécnica e de Computadores

Minimization of open orders using interval graphs

In this paper we address an order processing optimization problem known as the Minimization of Open Stacks Problem (MOSP). This problem consists in finding the best sequence for manufacturing the different products required by costumers, in a setting where only one product can be made at a time. The objective is to minimize the maximum number of incomplete orders from costumers that are being processed simultaneously. We present an integer programming model, based on the existence of a perfect elimination order in interval graphs, which finds an optimal sequence for the costumers orders. Among other economic advantages, manufacturing the products in this optimal sequence reduces the amount of space needed to store incomplete orders.

Domestic Consumption Simulation and Management Using a Continuous Consumption Management and Optimization Algorithm

The recent changes concerning the consumers’ active participation in the efficient management of load devices for one’s own interest and for the interest of the network operator, namely in the context of demand response, leads to the need for improved algorithms and tools. A continuous consumption optimization algorithm has been improved in order to better manage the shifted demand. It has been done in a simulation and user-interaction tool capable of being integrated in a multi-agent smart grid simulator already developed, and also capable of integrating several optimization algorithms to manage real and simulated loads. The case study of this paper enhances the advantages of the proposed algorithm and the benefits of using the developed simulation and user interaction tool.

A Learning Algorithm and System Approach to Address Exceptional Events in Domestic Consumption Management

The integration of the Smart Grid concept into the electric grid brings to the need for an active participation of small and medium players. This active participation can be achieved using decentralized decisions, in which the end consumer can manage loads regarding the Smart Grid needs. The management of loads must handle the users’ preferences, wills and needs. However, the users’ preferences, wills and needs can suffer changes when faced with exceptional events. This paper proposes the integration of exceptional events into the SCADA House Intelligent Management (SHIM) system developed by the authors, to handle machine learning issues in the domestic consumption context. An illustrative application and learning case study is provided in this paper.

Análise de desempenho de algoritmos de otimização para extração de parâmetros em modelos semicondutores de potência

Apresenta-se nesta tese uma revisão da literatura sobre a modelação de semicondutores de potência baseada na física e posterior análise de desempenho de dois métodos estocásticos, Particle Swarm Optimizaton (PSO) e Simulated Annealing (SA), quando utilizado para identificação eficiente de parâmetros de modelos de dispositivos semicondutores de potência, baseado na física. O conhecimento dos valores destes parâmetros, para cada dispositivo, é fundamental para uma simulação precisa do comportamento dinâmico do semicondutor. Os parâmetros são extraídos passo-a-passo durante simulação transiente e desempenham um papel relevante. Uma outra abordagem interessante nesta tese relaciona-se com o facto de que nos últimos anos, os métodos de modelação para dispositivos de potência têm emergido, com alta precisão e baixo tempo de execução baseado na Equação de Difusão Ambipolar (EDA) para díodos de potência e implementação no MATLAB numa estratégia de optimização formal. A equação da EDA é resolvida numericamente sob várias condições de injeções e o modelo é desenvolvido e implementado como um subcircuito no simulador IsSpice. Larguras de camada de depleção, área total do dispositivo, nível de dopagem, entre outras, são alguns dos parâmetros extraídos do modelo. Extração de parâmetros é uma parte importante de desenvolvimento de modelo. O objectivo de extração de parâmetros e otimização é determinar tais valores de parâmetros de modelo de dispositivo que minimiza as diferenças entre um conjunto de características medidas e resultados obtidos pela simulação de modelo de dispositivo. Este processo de minimização é frequentemente chamado de ajuste de características de modelos para dados de medição. O algoritmo implementado, PSO é uma técnica de heurística de otimização promissora, eficiente e recentemente proposta por Kennedy e Eberhart, baseado no comportamento social. As técnicas propostas são encontradas para serem robustas e capazes de alcançar uma solução que é caracterizada para ser precisa e global. Comparada com algoritmo SA já realizada, o desempenho da técnica proposta tem sido testado utilizando dados experimentais para extrair parâmetros de dispositivos reais das características I-V medidas. Para validar o modelo, comparação entre resultados de modelo desenvolvido com um outro modelo já desenvolvido são apresentados.

Design of New Materials for Passive Vibration Control

The aim of this contribution is to extend the techniques of composite materials design to non-linear material behaviour and apply it for design of new materials for passive vibration control. As a first step a computational tool allowing determination of macroscopic optimized one-dimensional isolator behaviour was developed. Voigt, Maxwell, standard and more complex material models can be implemented. Objective function considers minimization of the initial reaction and/or displacement peak as well as minimization of the steady-state amplitude of reaction and/or displacement. The complex stiffness approach is used to formulate the governing equations in an efficient way. Material stiffness parameters are assumed as non-linear functions of the displacement. The numerical solution is performed in the complex space. The steady-state solution in the complex space is obtained by an iterative process based on the shooting method which imposes the conditions of periodicity with respect to the known value of the period. Extension of the shooting method to the complex space is presented and verified. Non-linear behaviour of material parameters is then optimized by generic probabilistic meta-algorithm, simulated annealing. Dependence of the global optimum on several combinations of leading parameters of the simulated annealing procedure, like neighbourhood definition and annealing schedule, is also studied and analyzed. Procedure is programmed in MATLAB environment.

Geodupa - a geography-driven partial membership algorithm

Trabalho apresentado no âmbito do Mestrado em Engenharia Informática, como requisito parcial para obtenção do grau de Mestre em Engenharia Informática

Efficient Legendre spectral tau algorithm for solving two-sided space-time Caputo fractional advection-dispersion equation

In this paper we present the operational matrices of the left Caputo fractional derivative, right Caputo fractional derivative and Riemann–Liouville fractional integral for shifted Legendre polynomials. We develop an accurate numerical algorithm to solve the two-sided space–time fractional advection–dispersion equation (FADE) based on a spectral shifted Legendre tau (SLT) method in combination with the derived shifted Legendre operational matrices. The fractional derivatives are described in the Caputo sense. We propose a spectral SLT method, both in temporal and spatial discretizations for the two-sided space–time FADE. This technique reduces the two-sided space–time FADE to a system of algebraic equations that simplifies the problem. Numerical results carried out to confirm the spectral accuracy and efficiency of the proposed algorithm. By selecting relatively few Legendre polynomial degrees, we are able to get very accurate approximations, demonstrating the utility of the new approach over other numerical methods.

A filter algorithm: comparison with NLP solvers

The purpose of this work is to present an algorithm to solve nonlinear constrained optimization problems, using the filter method with the inexact restoration (IR) approach. In the IR approach two independent phases are performed in each iteration—the feasibility and the optimality phases. The first one directs the iterative process into the feasible region, i.e. finds one point with less constraints violation. The optimality phase starts from this point and its goal is to optimize the objective function into the satisfied constraints space. To evaluate the solution approximations in each iteration a scheme based on the filter method is used in both phases of the algorithm. This method replaces the merit functions that are based on penalty schemes, avoiding the related difficulties such as the penalty parameter estimation and the non-differentiability of some of them. The filter method is implemented in the context of the line search globalization technique. A set of more than two hundred AMPL test problems is solved. The algorithm developed is compared with LOQO and NPSOL software packages.

Algorithm for the Parkinson’s disease behavioural models characterization using a biosensor

Dissertação para obtenção do Grau de Mestre em Engenharia Biomédica

Chagas' disease: an algorithm for donor screening and positive donor counseling

Classical serological screening assays for Chagas' disease are time consuming and subjective. The objective of the present work is to evaluate the enzyme immuno-assay (ELISA) methodology and to propose an algorithm for blood banks to be applied to Chagas' disease. Seven thousand, nine hundred and ninety nine blood donor samples were screened by both reverse passive hemagglutination (RPHA) and indirect immunofluorescence assay (IFA). Samples reactive on RPHA and/or IFA were submitted to supplementary RPHA, IFA and complement fixation (CFA) tests. This strategy allowed us to create a panel of 60 samples to evaluate the ELISA methodology from 3 different manufacturers. The sensitivity of the screening by IFA and the 3 different ELISA's was 100%. The specificity was better on ELISA methodology. For Chagas disease, ELISA seems to be the best test for blood donor screening, because it showed high sensitivity and specificity, it is not subjective and can be automated. Therefore, it was possible to propose an algorithm to screen samples and confirm donor results at the blood bank.

Diffusional kurtosis imaging using a fast heuristic constrained linear least squares algorithm: a plugin for OsiriX

Diffusion Kurtosis Imaging (DKI) is a fairly new magnetic resonance imag-ing (MRI) technique that tackles the non-gaussian motion of water in biological tissues by taking into account the restrictions imposed by tissue microstructure, which are not considered in Diffusion Tensor Imaging (DTI), where the water diffusion is considered purely gaussian. As a result DKI provides more accurate information on biological structures and is able to detect important abnormalities which are not visible in standard DTI analysis. This work regards the development of a tool for DKI computation to be implemented as an OsiriX plugin. Thus, as OsiriX runs under Mac OS X, the pro-gram is written in Objective-C and also makes use of Apple’s Cocoa framework. The whole program is developed in the Xcode integrated development environ-ment (IDE). The plugin implements a fast heuristic constrained linear least squares al-gorithm (CLLS-H) for estimating the diffusion and kurtosis tensors, and offers the user the possibility to choose which maps are to be generated for not only standard DTI quantities such as Mean Diffusion (MD), Radial Diffusion (RD), Axial Diffusion (AD) and Fractional Anisotropy (FA), but also DKI metrics, Mean Kurtosis (MK), Radial Kurtosis (RK) and Axial Kurtosis (AK).The plugin was subjected to both a qualitative and a semi-quantitative analysis which yielded convincing results. A more accurate validation pro-cess is still being developed, after which, and with some few minor adjust-ments the plugin shall become a valid option for DKI computation

EDP - A new estimation algorithm

The aim of this work project is to analyze the current algorithm used by EDP to estimate their clients’ electrical energy consumptions, create a new algorithm and compare the advantages and disadvantages of both. This new algorithm is different from the current one as it incorporates some effects from temperature variations. The results of the comparison show that this new algorithm with temperature variables performed better than the same algorithm without temperature variables, although there is still potential for further improvements of the current algorithm, if the prediction model is estimated using a sample of daily data, which is the case of the current EDP algorithm.