Vertex component analysis: a geometric-based approach to unmix hyperspectral data

Hyperspectral remote sensing exploits the electromagnetic scattering patterns of the different materials at specific wavelengths [2, 3]. Hyperspectral sensors have been developed to sample the scattered portion of the electromagnetic spectrum extending from the visible region through the near-infrared and mid-infrared, in hundreds of narrow contiguous bands [4, 5]. The number and variety of potential civilian and military applications of hyperspectral remote sensing is enormous [6, 7]. Very often, the resolution cell corresponding to a single pixel in an image contains several substances (endmembers) [4]. In this situation, the scattered energy is a mixing of the endmember spectra. A challenging task underlying many hyperspectral imagery applications is then decomposing a mixed pixel into a collection of reflectance spectra, called endmember signatures, and the corresponding abundance fractions [8–10]. Depending on the mixing scales at each pixel, the observed mixture is either linear or nonlinear [11, 12]. Linear mixing model holds approximately when the mixing scale is macroscopic [13] and there is negligible interaction among distinct endmembers [3, 14]. If, however, the mixing scale is microscopic (or intimate mixtures) [15, 16] and the incident solar radiation is scattered by the scene through multiple bounces involving several endmembers [17], the linear model is no longer accurate. Linear spectral unmixing has been intensively researched in the last years [9, 10, 12, 18–21]. It considers that a mixed pixel is a linear combination of endmember signatures weighted by the correspondent abundance fractions. Under this model, and assuming that the number of substances and their reflectance spectra are known, hyperspectral unmixing is a linear problem for which many solutions have been proposed (e.g., maximum likelihood estimation [8], spectral signature matching [22], spectral angle mapper [23], subspace projection methods [24,25], and constrained least squares [26]). In most cases, the number of substances and their reflectances are not known and, then, hyperspectral unmixing falls into the class of blind source separation problems [27]. Independent component analysis (ICA) has recently been proposed as a tool to blindly unmix hyperspectral data [28–31]. ICA is based on the assumption of mutually independent sources (abundance fractions), which is not the case of hyperspectral data, since the sum of abundance fractions is constant, implying statistical dependence among them. This dependence compromises ICA applicability to hyperspectral images as shown in Refs. [21, 32]. In fact, ICA finds the endmember signatures by multiplying the spectral vectors with an unmixing matrix, which minimizes the mutual information among sources. If sources are independent, ICA provides the correct unmixing, since the minimum of the mutual information is obtained only when sources are independent. This is no longer true for dependent abundance fractions. Nevertheless, some endmembers may be approximately unmixed. These aspects are addressed in Ref. [33]. Under the linear mixing model, the observations from a scene are in a simplex whose vertices correspond to the endmembers. Several approaches [34–36] have exploited this geometric feature of hyperspectral mixtures [35]. Minimum volume transform (MVT) algorithm [36] determines the simplex of minimum volume containing the data. The method presented in Ref. [37] is also of MVT type but, by introducing the notion of bundles, it takes into account the endmember variability usually present in hyperspectral mixtures. The MVT type approaches are complex from the computational point of view. Usually, these algorithms find in the first place the convex hull defined by the observed data and then fit a minimum volume simplex to it. For example, the gift wrapping algorithm [38] computes the convex hull of n data points in a d-dimensional space with a computational complexity of O(nbd=2cþ1), where bxc is the highest integer lower or equal than x and n is the number of samples. The complexity of the method presented in Ref. [37] is even higher, since the temperature of the simulated annealing algorithm used shall follow a log( ) law [39] to assure convergence (in probability) to the desired solution. Aiming at a lower computational complexity, some algorithms such as the pixel purity index (PPI) [35] and the N-FINDR [40] still find the minimum volume simplex containing the data cloud, but they assume the presence of at least one pure pixel of each endmember in the data. 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. PPI algorithm uses the minimum noise fraction (MNF) [41] as a preprocessing step to reduce dimensionality and to improve the signal-to-noise ratio (SNR). The algorithm then projects every spectral vector onto skewers (large number of random vectors) [35, 42,43]. The points corresponding to extremes, for each skewer direction, are stored. A cumulative account records the number of times each pixel (i.e., a given spectral vector) is found to be an extreme. The pixels with the highest scores are the purest ones. N-FINDR algorithm [40] is based on the fact that in p spectral dimensions, the p-volume defined by a simplex formed by the purest pixels is larger than any other volume defined by any other combination of pixels. This algorithm finds the set of pixels defining the largest volume by inflating a simplex inside the data. ORA SIS [44, 45] is a hyperspectral framework developed by the U.S. Naval Research Laboratory consisting of several algorithms organized in six modules: exemplar selector, adaptative learner, demixer, knowledge base or spectral library, and spatial postrocessor. The first step consists in flat-fielding the spectra. Next, the exemplar selection module is used to select spectral vectors that best represent the smaller convex cone containing the data. The other pixels are rejected when the spectral angle distance (SAD) is less than a given thresh old. The procedure finds the basis for a subspace of a lower dimension using a modified Gram–Schmidt orthogonalizati on. The selected vectors are then projected onto this subspace and a simplex is found by an MV T pro cess. ORA SIS is oriented to real-time target detection from uncrewed air vehicles using hyperspectral data [46]. In this chapter we develop a new algorithm to unmix linear mixtures of endmember spectra. First, the algorithm determines the number of endmembers and the signal subspace using a newly developed concept [47, 48]. Second, the algorithm extracts the most pure pixels present in the data. Unlike other methods, this algorithm is completely automatic and unsupervised. To estimate the number of endmembers and the signal subspace in hyperspectral linear mixtures, the proposed scheme begins by estimating sign al and noise correlation matrices. The latter is based on multiple regression theory. The signal subspace is then identified by selectin g the set of signal eigenvalue s that best represents the data, in the least-square sense [48,49 ], we note, however, that VCA works with projected and with unprojected data. The extraction of the end members exploits two facts: (1) the endmembers are the vertices of a simplex and (2) the affine transformation of a simplex is also a simplex. As PPI and N-FIND R algorithms, VCA also assumes the presence of pure pixels in the data. The algorithm iteratively projects data on to a direction orthogonal to the subspace spanned by the endmembers already determined. The new end member signature corresponds to the extreme of the projection. The algorithm iterates until all end members are exhausted. VCA performs much better than PPI and better than or comparable to N-FI NDR; yet it has a computational complexity between on e and two orders of magnitude lower than N-FINDR. The chapter is structure d as follows. Section 19.2 describes the fundamentals of the proposed method. Section 19.3 and Section 19.4 evaluate the proposed algorithm using simulated and real data, respectively. Section 19.5 presents some concluding remarks.

Random amplified polymorphic DNA (RAPD) analysis of Lutzomyia longipalpis laboratory populations

The phlebotomine sand fly Lutzomyia longipalpis has been incriminated as a vector of American visceral leishmaniasis, caused by Leishmania chagasi. However, some evidence has been accumulated suggesting that it may exist in nature not as a single but as a species complex. Our goal was to compare four laboratory reference populations of L. longipalpis from distinct geographic regions at the molecular level by RAPD-PCR. We screened genomic DNA for polymorphic sites by PCR amplification with decamer single primers of arbitrary nucleotide sequences. One primer distinguished one population (Marajó Island, Pará State, Brazil) from the other three (Lapinha Cave, Minas Gerais State, Brazil; Melgar, Tolima Department, Colombia and Liberia, Guanacaste Province, Costa Rica). The population-specific and the conserved RAPD-PCR amplified fragments were cloned and shown to differ only in number of internal repeats.

Full analysis of suspension geometry and chassis performance of Formula Student racing car

Formula Student events gather engineering students, who compete, designing, building and racing single-seater cars. The team of ISEP is working on its first car that soon will take part in this competition. This work aims to analyze the current design’s chassis, focusing on suspension geometry and frame’s performance. After analyzing results of the tests planned suggestions, that can be taken into consideration during design process of next cars will be presented. As the car has not been tested yet this work can also be helpful to explain its performance on the track later.

Models and tools for value systems analysis in collaborative environments

Dissertation to obtain the degree of Doctor in Electrical and Computer Engineering, specialization of Collaborative Networks

New modified electrochemical conductive paper support for BSA detection

Chemical sensors and biosensors are widely used to detect various kinds of protein target biomolecules. Molecularly Imprinted Polymers (MIPs) have raised great interest in this area, because these act as antibody-like recognition materials, with high affinity to the template molecule. Compared to natural antibodies, these are also of lower cost and higher stability. There are different types of supports used to carry MIP materials, mostly of these made of gold, favourably assembled on a Screen Printed Electrode (SPE) strategy. For this work a new kind of support for the sensing layer was developed: conductive paper. This support was made by modifying first cellulose paper with paraffin wax (to make it waterproof), and casting a carbon-ink on it afterwards, to turn it conductive. The SPAM approach previously reported in1 was employed herein to assemble to MIP sensing material on the conductive paper. The selected charged monomers were (vinylbenzyl) trimethlammonium chloride (positive charge) or vinylbenzoic acid (negative charge), used to generate binding positions with single-type charge (positive or negative). The non-specific binding area of the MIP layer was assembled by chronoamperometry-assisted polymerization (at 1 V, for 60, 120 or 180 seconds) of vinylbenzoate, cross-linked with ethylene glycol vinyl ether. The BSA biomolecules lying within the polymeric matrix were removed by Proteinase K action. All preparation stages of the MIP assembly were followed by FTIR, Raman spectroscopy and, electrochemical analysis. In general, the best results were obtained for longer polymerization times and positively charged binding sites (which was consistent with a negatively-charged protein under physiological pH, as BSA). Linear responses against BSA concentration ranged from 0.005 to 100 mg/mL, in PBS buffer standard solutions. The sensor was further calibrated in standard solutions that were prepared in synthetic or real urine, and the analytical response became more sensitive and stable. Compared to the literature, the detection capability of the developed device is better than most of the reported electrodes. Overall, the simplicity, low cost and good analytical performance of the BSA SPE device, prepared with positively charged binding positions, seems a suitable approach for practical application in clinical context. Further studies with real samples are required, as well as gathering with electronic-supporting devices to allow on-site readings.

Recycling old screen-printed electrodes with newly designed plastic antibodies on the wall of carbon nanotubes as sensory element for in situ detection of bacterial toxins in water

Using low cost portable devices that enable a single analytical step for screening environmental contaminants is today a demanding issue. This concept is here tried out by recycling screen-printed electrodes that were to be disposed of and by choosing as sensory element a low cost material offering specific response for an environmental contaminant. Microcystins (MCs) were used as target analyte, for being dangerous toxins produced by cyanobacteria released into water bodies. The sensory element was a plastic antibody designed by surface imprinting with carefully selected monomers to ensure a specific response. These were designed on the wall of carbon nanotubes, taking advantage of their exceptional electrical properties. The stereochemical ability of the sensory material to detect MCs was checked by preparing blank materials where the imprinting stage was made without the template molecule. The novel sensory material for MCs was introduced in a polymeric matrix and evaluated against potentiometric measurements. Nernstian response was observed from 7.24 × 10−10 to 1.28 × 10−9 M in buffer solution (10 mM HEPES, 150 mM NaCl, pH 6.6), with average slopes of −62 mVdecade−1 and detection capabilities below 1 nM. The blank materials were unable to provide a linear response against log(concentration), showing only a slight potential change towards more positive potentials with increasing concentrations (while that ofthe plastic antibodies moved to more negative values), with a maximum rate of +33 mVdecade−1. The sensors presented good selectivity towards sulphate, iron and ammonium ions, and also chloroform and tetrachloroethylene (TCE) and fast response (<20 s). This concept was successfully tested on the analysis of spiked environmental water samples. The sensors were further applied onto recycled chips, comprehending one site for the reference electrode and two sites for different selective membranes, in a biparametric approach for “in situ” analysis.

Label-free Detection of Microcystin-LR in Waters Using Real-Time Potentiometric Biosensors Based on Single-Walled Carbon Nanotubes Imprinted Polymers

Microcystin-LR (MC-LR) is a dangerous toxin found in environmental waters, quantified by high performance liquid chromatography and/or enzyme-linked immunosorbent assays. Quick, low cost and on-site analysis is thus required to ensure human safety and wide screening programs. This work proposes label-free potentiometric sensors made of solid-contact electrodes coated with a surface imprinted polymer on the surface of Multi-Walled Carbon NanoTubes (CNTs) incorporated in a polyvinyl chloride membrane. The imprinting effect was checked by using non-imprinted materials. The MC-LR sensitive sensors were evaluated, characterized and applied successfully in spiked environmental waters. The presented method offered the advantages of low cost, portability, easy operation and suitability for adaptation to flow methods.

Novel LTCC-potentiometric microfluidic device for biparametric analysis of organic compounds carrying plastic antibodies as ionophores: Application to sulfamethoxazole and trimethoprim

Monitoring organic environmental contaminants is of crucial importance to ensure public health. This requires simple, portable and robust devices to carry out on-site analysis. For this purpose, a low-temperature co-fired ceramics (LTCC) microfluidic potentiometric device (LTCC/μPOT) was developed for the first time for an organic compound: sulfamethoxazole (SMX). Sensory materials relied on newly designed plastic antibodies. Sol–gel, self-assembling monolayer and molecular-imprinting techniques were merged for this purpose. Silica beads were amine-modified and linked to SMX via glutaraldehyde modification. Condensation polymerization was conducted around SMX to fill the vacant spaces. SMX was removed after, leaving behind imprinted sites of complementary shape. The obtained particles were used as ionophores in plasticized PVC membranes. The most suitable membrane composition was selected in steady-state assays. Its suitability to flow analysis was verified in flow-injection studies with regular tubular electrodes. The LTCC/μPOT device integrated a bidimensional mixer, an embedded reference electrode based on Ag/AgCl and an Ag-based contact screen-printed under a micromachined cavity of 600 μm depth. The sensing membranes were deposited over this contact and acted as indicating electrodes. Under optimum conditions, the SMX sensor displayed slopes of about −58.7 mV/decade in a range from 12.7 to 250 μg/mL, providing a detection limit of 3.85 μg/mL and a sampling throughput of 36 samples/h with a reagent consumption of 3.3 mL per sample. The system was adjusted later to multiple analyte detection by including a second potentiometric cell on the LTCC/μPOT device. No additional reference electrode was required. This concept was applied to Trimethoprim (TMP), always administered concomitantly with sulphonamide drugs, and tested in fish-farming waters. The biparametric microanalyzer displayed Nernstian behaviour, with average slopes −54.7 (SMX) and +57.8 (TMP) mV/decade. To demonstrate the microanalyzer capabilities for real applications, it was successfully applied to single and simultaneous determination of SMX and TMP in aquaculture waters.

Integer and Fractional Order Entropy Analysis of Earthquake Data-series

This paper studies the statistical distributions of worldwide earthquakes from year 1963 up to year 2012. A Cartesian grid, dividing Earth into geographic regions, is considered. Entropy and the Jensen–Shannon divergence are used to analyze and compare real-world data. Hierarchical clustering and multi-dimensional scaling techniques are adopted for data visualization. Entropy-based indices have the advantage of leading to a single parameter expressing the relationships between the seismic data. Classical and generalized (fractional) entropy and Jensen–Shannon divergence are tested. The generalized measures lead to a clear identification of patterns embedded in the data and contribute to better understand earthquake distributions.

Entropy Analysis of Industrial Accident Data Series

Complex industrial plants exhibit multiple interactions among smaller parts and with human operators. Failure in one part can propagate across subsystem boundaries causing a serious disaster. This paper analyzes the industrial accident data series in the perspective of dynamical systems. First, we process real world data and show that the statistics of the number of fatalities reveal features that are well described by power law (PL) distributions. For early years, the data reveal double PL behavior, while, for more recent time periods, a single PL fits better into the experimental data. Second, we analyze the entropy of the data series statistics over time. Third, we use the Kullback–Leibler divergence to compare the empirical data and multidimensional scaling (MDS) techniques for data analysis and visualization. Entropy-based analysis is adopted to assess complexity, having the advantage of yielding a single parameter to express relationships between the data. The classical and the generalized (fractional) entropy and Kullback–Leibler divergence are used. The generalized measures allow a clear identification of patterns embedded in the data.

A Generic and Compositional Framework for Multicore Response Time Analysis

23rd International Conference on Real-Time Networks and Systems (RTNS 2015). 4 to 6, Nov, 2015, Main Track. Lille, France. Best Paper Award Nominee

Heterodimeric nitrate reductase (NapAB) from Cupriavidus necator H16: purification, crystallization and preliminary X-ray analysis

Acta Cryst. (2007). F63, 516–519

Determinação da tenacidade à fratura de adesivos estruturais em modo misto pelo ensaio Single-Leg Bending

A utilização de juntas adesivas em aplicações industriais tem vindo a aumentar, em detrimento dos métodos tradicionais tais como a soldadura, brasagem e ligações aparafusadas e rebitadas. Este facto deve-se às vantagens que estas oferecem, como o facto de serem mais leves, comportarem-se bem sob cargas cíclicas ou de fadiga, a ligação de materiais diferentes e menores concentrações de tensões. Para aumentar a confiança no projeto de estruturas adesivas, é importante conseguir prever com precisão a sua resistência mecânica e respetivas propriedades de fratura (taxa crítica de libertação de energia de deformação à tração, GIC, e corte, GIIC). Estas propriedades estão diretamente relacionadas com a Mecânica da Fratura e são estimadas através de uma análise energética. Para este efeito, distinguem-se três tipos de modelos: modelos que necessitam da medição do comprimento de fenda durante a propagação do dano, modelos que utilizam um comprimento de fenda equivalente e métodos baseados no integral J. Como na maioria dos casos as solicitações ocorrem em modo misto (combinação de tração com corte), é de grande importância a perceção da fratura nesta condições, nomeadamente das taxas de libertação de energia relativamente a diferentes critérios ou envelopes de fratura. Esta comparação permite, por exemplo, averiguar qual o melhor critério energético de rotura a utilizar em modelos numéricos baseados em Modelos de Dano Coesivo. Neste trabalho é realizado um estudo experimental utilizando o ensaio Single-Leg Bending (SLB) em provetes colados com três tipos de adesivos, de forma a estudar e comparar as suas propriedades de fratura. Para tal, são aplicados alguns modelos de redução da taxa de libertação de energia de deformação à tração, GI, e corte, GII, enquadrados nos modelos que necessitam da medição do comprimento de fenda e nos modelos que utilizam um comprimento de fenda equivalente. Numa fase posterior, procedeu-se à análise e comparação dos resultados adquiridos durante a fase experimental de GI e GII de cada adesivo. A discussão de resultados foi também feita através da análise dos valores obtidos em diversos envelopes de fratura, no sentido de averiguar qual o critério de rotura mais adequado a considerar para cada adesivo. Foi obtida uma concordância bastante boa entre métodos de determinação de GI e GII, com exceção do adesivo mais dúctil, para o qual o método baseado no comprimento de fenda equivalente apresentou resultados ligeiramente superiores.

Towards realistic core-failure-resilient scheduling and analysis

Presented at IEEE Real-Time Systems Symposium (RTSS 2015). 1 to 4, Dec, 2015. San Antonio, U.S.A..

Adhesive selection for single lap bonded joints: Experimentation and advanced techniques for strength prediction

The integrity of multi-component structures is usually determined by their unions. Adhesive-bonding is often used over traditional methods because of the reduction of stress concentrations, reduced weight penalty, and easy manufacturing. Commercial adhesives range from strong and brittle (e.g., Araldite® AV138) to less strong and ductile (e.g., Araldite® 2015). A new family of polyurethane adhesives combines high strength and ductility (e.g., Sikaforce® 7888). In this work, the performance of the three above-mentioned adhesives was tested in single lap joints with varying values of overlap length (LO). The experimental work carried out is accompanied by a detailed numerical analysis by finite elements, either based on cohesive zone models (CZM) or the extended finite element method (XFEM). This procedure enabled detailing the performance of these predictive techniques applied to bonded joints. Moreover, it was possible to evaluate which family of adhesives is more suited for each joint geometry. CZM revealed to be highly accurate, except for largely ductile adhesives, although this could be circumvented with a different cohesive law. XFEM is not the most suited technique for mixed-mode damage growth, but a rough prediction was achieved.