FPGA-based architecture for hyperspectral unmixing

This paper proposes an FPGA-based architecture for onboard hyperspectral unmixing. This method based on the Vertex Component Analysis (VCA) has several advantages, namely it is unsupervised, fully automatic, and it works without dimensionality reduction (DR) pre-processing step. The architecture has been designed for a low cost Xilinx Zynq board with a Zynq-7020 SoC FPGA based on the Artix-7 FPGA programmable logic and tested using real hyperspectral datasets. Experimental results indicate that the proposed implementation can achieve real-time processing, while maintaining the methods accuracy, which indicate the potential of the proposed platform to implement high-performance, low cost embedded systems.

Parallel hyperspectral unmixing method via split augmented lagrangian on GPU

One of the main problems of hyperspectral data analysis is the presence of mixed pixels due to the low spatial resolution of such images. Linear spectral unmixing aims at inferring pure spectral signatures and their fractions at each pixel of the scene. The huge data volumes acquired by hyperspectral sensors put stringent requirements on processing and unmixing methods. This letter proposes an efficient implementation of the method called simplex identification via split augmented Lagrangian (SISAL) which exploits the graphics processing unit (GPU) architecture at low level using Compute Unified Device Architecture. SISAL aims to identify the endmembers of a scene, i.e., is able to unmix hyperspectral data sets in which the pure pixel assumption is violated. The proposed implementation is performed in a pixel-by-pixel fashion using coalesced accesses to memory and exploiting shared memory to store temporary data. Furthermore, the kernels have been optimized to minimize the threads divergence, therefore achieving high GPU occupancy. The experimental results obtained for the simulated and real hyperspectral data sets reveal speedups up to 49 times, which demonstrates that the GPU implementation can significantly accelerate the method's execution over big data sets while maintaining the methods accuracy.

Hyperspectral signal subspace estimation

Given an hyperspectral image, the determination of the number of endmembers and the subspace where they live without any prior knowledge is crucial to the success of hyperspectral image analysis. This paper introduces a new minimum mean squared error based approach to infer the signal subspace in hyperspectral imagery. The method, termed hyperspectral signal identification by minimum error (HySime), is eigendecomposition based and it does not depend on any tuning parameters. It first estimates the signal and noise correlation matrices and then selects the subset of eigenvalues that best represents the signal subspace in the least squared error sense. 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.

Parallel sparse unmixing of hyperspectral data

In this paper, a new parallel method for sparse spectral unmixing of remotely sensed hyperspectral data on commodity graphics processing units (GPUs) is presented. A semi-supervised approach is adopted, which relies on the increasing availability of spectral libraries of materials measured on the ground instead of resorting to endmember extraction methods. This method is based on the spectral unmixing by splitting and augmented Lagrangian (SUNSAL) that estimates the material's abundance fractions. The parallel method is performed in a pixel-by-pixel fashion and its implementation properly exploits the GPU architecture at low level, thus taking full advantage of the computational power of GPUs. Experimental results obtained for simulated and real hyperspectral datasets reveal significant speedup factors, up to 1 64 times, with regards to optimized serial implementation.

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 indústria de defesa na cooperação Brasil-Portugal

The global dynamics of alliances are strongly determined by the level of cooperation among states. This cooperation can be embodied in various aspects, but the level of defense and security cooperation becomes usually more doctrinal and lasting. By the nature of sovereignty that instills in the bilateral relationship, cooperation at defense and security level can leverages other forms of cooperation. The circumstances and relational balance between Brazil and Portugal seem to evolve towards distancing opportunities, despite they are culturally and institutionally untainted. The economic dynamics, the strategic projection in global sustainability terms, the scale and ambition of Brazilian regional leadership, contrasts with the actual context of Portugal, distancing himself both on the stage where they operate. On the other hand, the historical and cultural roots, the language, the affinity of the peoples of CPLP and some opportunities for economic niches, trend to attract both countries. The condition of Portugal in NATO and Europe, coupled with the ability to export technical and human resources to value-added for Brazil, seems also to become approaching factors. On the balance of these dynamics, there is a set of exogenous factors (economic, external global relations matrix, regional stability, among others), which are not always controlled by any of both countries. These factors call for strong capacity for foresight analysis and decision making, with the inherent risk. There is cooperation vectors that are not apparently penalized by geographic distance, or by the difference of realities. Among these vectors we shall highlight synergies in technological niches, highly tradable goods and, mostly, using the domain of dual technologies. The thirteen niches herein identified are: Monitoring, Navigation, Command and Control, Electronics, Optoelectronics, Communication and remote sensing, Information Technologies, Flight Simulation, Specialized Training, Fiber Optic Sensors, Materials Engineering, Nanotechnology and Communications. Cumulating with identified opportunities in traditional relational framework, both countries are growing (in geography and economic terms) into the Atlantic, making it a central element in the bilateral approach. By being at the same time a growing stage of disputes and which stability tends to be threatened, it will be done an analysis of these synergistic vectors, superimposed on the impact on Atlantic securitization process.

Prospecção de ocorrências filonianas do tipo LCT nas formações metassedimentares entre Boticas e Montalegre (N de Portugal)

Dissertação de mestrado em Geologia (área de especialização em Valorização de Recursos Geológicos)

Urban image classification with semisupervised multiscale cluster kernels

This paper presents a semisupervised support vector machine (SVM) that integrates the information of both labeled and unlabeled pixels efficiently. Method's performance is illustrated in the relevant problem of very high resolution image classification of urban areas. The SVM is trained with the linear combination of two kernels: a base kernel working only with labeled examples is deformed by a likelihood kernel encoding similarities between labeled and unlabeled examples. Results obtained on very high resolution (VHR) multispectral and hyperspectral images show the relevance of the method in the context of urban image classification. Also, its simplicity and the few parameters involved make the method versatile and workable by unexperienced users.

Active Learning of Very-High Resolution Optical Imagery with SVM: Entropy vs Margin Sampling

An active learning method is proposed for the semi-automatic selection of training sets in remote sensing image classification. The method adds iteratively to the current training set the unlabeled pixels for which the prediction of an ensemble of classifiers based on bagged training sets show maximum entropy. This way, the algorithm selects the pixels that are the most uncertain and that will improve the model if added in the training set. The user is asked to label such pixels at each iteration. Experiments using support vector machines (SVM) on an 8 classes QuickBird image show the excellent performances of the methods, that equals accuracies of both a model trained with ten times more pixels and a model whose training set has been built using a state-of-the-art SVM specific active learning method

Decision fusion for the classification of hyperspectral data: Outcome of the 2008 GRS-S data fusion contest

The 2008 Data Fusion Contest organized by the IEEE Geoscience and Remote Sensing Data Fusion Technical Committee deals with the classification of high-resolution hyperspectral data from an urban area. Unlike in the previous issues of the contest, the goal was not only to identify the best algorithm but also to provide a collaborative effort: The decision fusion of the best individual algorithms was aiming at further improving the classification performances, and the best algorithms were ranked according to their relative contribution to the decision fusion. This paper presents the five awarded algorithms and the conclusions of the contest, stressing the importance of decision fusion, dimension reduction, and supervised classification methods, such as neural networks and support vector machines.

Multi-modal Change Detection, Application to the Detection of Flooded Areas: Outcome of the 2009-2010 data fusion contest

The 2009-2010 Data Fusion Contest organized by the Data Fusion Technical Committee of the IEEE Geoscience and Remote Sensing Society was focused on the detection of flooded areas using multi-temporal and multi-modal images. Both high spatial resolution optical and synthetic aperture radar data were provided. The goal was not only to identify the best algorithms (in terms of accuracy), but also to investigate the further improvement derived from decision fusion. This paper presents the four awarded algorithms and the conclusions of the contest, investigating both supervised and unsupervised methods and the use of multi-modal data for flood detection. Interestingly, a simple unsupervised change detection method provided similar accuracy as supervised approaches, and a digital elevation model-based predictive method yielded a comparable projected change detection map without using post-event data.

Multisensor change detection with nonlinear canonical correlation

The analysis of multi-modal and multi-sensor images is nowadays of paramount importance for Earth Observation (EO) applications. There exist a variety of methods that aim at fusing the different sources of information to obtain a compact representation of such datasets. However, for change detection existing methods are often unable to deal with heterogeneous image sources and very few consider possible nonlinearities in the data. Additionally, the availability of labeled information is very limited in change detection applications. For these reasons, we present the use of a semi-supervised kernel-based feature extraction technique. It incorporates a manifold regularization accounting for the geometric distribution and jointly addressing the small sample problem. An exhaustive example using Landsat 5 data illustrates the potential of the method for multi-sensor change detection.

Aplicação dos Sistemas de Informação Geográfica em Estudos de Caracterização Biofísica da Ilha de Santo Antão em Cabo Verde

Com características morfológicas e edafo-climáticas extremamente diversificadas, a ilha de Santo Antão em Cabo Verde apresenta uma reconhecida vulnerabilidade ambiental a par de uma elevada carência de estudos científicos que incidam sobre essa realidade e sirvam de base à uma compreensão integrada dos fenómenos. A cartografia digital e as tecnologias de informação geográfica vêm proporcionando um avanço tecnológico na colecção, armazenamento e processamento de dados espaciais. Várias ferramentas actualmente disponíveis permitem modelar uma multiplicidade de factores, localizar e quantificar os fenómenos bem como e definir os níveis de contribuição de diferentes factores no resultado final. No presente estudo, desenvolvido no âmbito do curso de pós-graduação e mestrado em sistemas de Informação geográfica realizado pela Universidade de Trás-os-Montes e Alto Douro, pretende-se contribuir para a minimização do deficit de informação relativa às características biofísicas da citada ilha, recorrendo-se à aplicação de tecnologias de informação geográfica e detecção remota, associadas à análise estatística multivariada. Nesse âmbito, foram produzidas e analisadas cartas temáticas e desenvolvido um modelo de análise integrada de dados. Com efeito, a multiplicidade de variáveis espaciais produzidas, de entre elas 29 variáveis com variação contínua passíveis de influenciar as características biofísicas da região e, possíveis ocorrências de efeitos mútuos antagónicos ou sinergéticos, condicionam uma relativa complexidade à interpretação a partir dos dados originais. Visando contornar este problema, recorre-se a uma rede de amostragem sistemática, totalizando 921 pontos ou repetições, para extrair os dados correspondentes às 29 variáveis nos pontos de amostragem e, subsequente desenvolvimento de técnicas de análise estatística multivariada, nomeadamente a análise em componentes principais. A aplicação destas técnicas permitiu simplificar e interpretar as variáreis originais, normalizando-as e resumindo a informação contida na diversidade de variáveis originais, correlacionadas entre si, num conjunto de variáveis ortogonais (não correlacionadas), e com níveis de importância decrescente, as componentes principais. Fixou-se como meta a concentração de 75% da variância dos dados originais explicadas pelas primeiras 3 componentes principais e, desenvolveu-se um processo interactivo em diferentes etapas, eliminando sucessivamente as variáveis menos representativas. Na última etapa do processo as 3 primeiras CP resultaram em 74,54% da variância dos dados originais explicadas mas, que vieram a demonstrar na fase posterior, serem insuficientes para retratar a realidade. Optou-se pela inclusão da 4ª CP (CP4), com a qual 84% da referida variância era explicada e, representando oito variáveis biofísicas: a altitude, a densidade hidrográfica, a densidade de fracturação geológica, a precipitação, o índice de vegetação, a temperatura, os recursos hídricos e a distância à rede hidrográfica. A subsequente interpolação da 1ª componente principal (CP1) e, das principais variáveis associadas as componentes CP2, CP3 e CP4 como variáveis auxiliares, recorrendo a técnicas geoestatística em ambiente ArcGIS permitiu a obtenção de uma carta representando 84% da variação das características biofísicas no território. A análise em clusters validada pelo teste “t de Student” permitiu reclassificar o território em 6 unidades biofísicas homogéneas. Conclui-se que, as tecnologias de informação geográfica actualmente disponíveis a par de facilitar análises interactivas e flexíveis, possibilitando que se faça variar temas e critérios, integrar novas informações e introduzir melhorias em modelos construídos com bases em informações disponíveis num determinado contexto, associadas a técnicas de análise estatística multivariada, possibilitam, com base em critérios científicos, desenvolver a análise integrada de múltiplas variáveis biofísicas cuja correlação entre si, torna complexa a compreensão integrada dos fenómenos.

Kernel-based mapping of orographic rainfall enhancement in the Swiss Alps as detected by weatherradar

In this paper, we develop a data-driven methodology to characterize the likelihood of orographic precipitation enhancement using sequences of weather radar images and a digital elevation model (DEM). Geographical locations with topographic characteristics favorable to enforce repeatable and persistent orographic precipitation such as stationary cells, upslope rainfall enhancement, and repeated convective initiation are detected by analyzing the spatial distribution of a set of precipitation cells extracted from radar imagery. Topographic features such as terrain convexity and gradients computed from the DEM at multiple spatial scales as well as velocity fields estimated from sequences of weather radar images are used as explanatory factors to describe the occurrence of localized precipitation enhancement. The latter is represented as a binary process by defining a threshold on the number of cell occurrences at particular locations. Both two-class and one-class support vector machine classifiers are tested to separate the presumed orographic cells from the nonorographic ones in the space of contributing topographic and flow features. Site-based validation is carried out to estimate realistic generalization skills of the obtained spatial prediction models. Due to the high class separability, the decision function of the classifiers can be interpreted as a likelihood or susceptibility of orographic precipitation enhancement. The developed approach can serve as a basis for refining radar-based quantitative precipitation estimates and short-term forecasts or for generating stochastic precipitation ensembles conditioned on the local topography.