Made available in DSpace on 2018-08-01T21:36:53Z (GMT). No. of bitstreams: 1 tese_10765_Dissertação_final_Rayza.pdf: 2180620 bytes, checksum: 5e88e543a5adef7347618d2eef3cfb62 (MD5) Previous issue date: 2017-03-27 / A calibração multivariada associada à técnica de infravermelho é uma alternativa aos métodos tradicionais para determinação de parâmetros físico-químicos em petróleo. Entretanto, o modelo multivariado construído é aplicável somente para o instrumento no qual os espectros foram obtidos. A transferência de calibração de modelos multivariados é de fundamental importância para a indústria petrolífera, levando-se em conta que aumenta a aplicabilidade de modelos e permite estimar com rapidez e com poucos gastos diversas propriedades físico-químicas. Este trabalho foi dividido em duas seções: a primeira é dedicada à avaliação modelos de transferência entre dois instrumentos de infravermelho a região média (MIR) para modelos PLS (mínimos quadrados parciais) e OPLS (projeções ortogonais em estruturas latentes); e a segunda visa aplicar a transferência PDS (padronização direta por partes) a modelos PLS para predição de densidade API, TIAC (temperatura de início de aparecimento de cristais), NAT (número de acidez total) e NAN (número de acidez naftênica). Dentre os métodos quimiométricos SBC (correção de declive e viés), FR (recalibração total), DS (padronização direta) e PDS avaliados para previsão de densidade API, o modelo de transferência PDS aplicado ao modelo OPLS resultou na melhor capacidade preditiva (RMSEP de 1,48). A aplicação direta de PDS a um modelo PLS original conta com a vantagem de aproveitar modelos já consolidados, e os resultados da primeira seção indicam que isto é uma possibilidade. Na segunda seção, as propriedades API, TIAC, NAT e NAN foram modeladas e validadas por PLS para o instrumento primário. Espectros secundários, diferentes daqueles da primeira seção, passaram por uma interpolação polinomial a fim de igualar as variáveis às do instrumento primário, antes da aplicação da transferência PDS. Os espectros transferidos e processados por airPLS, técnica iterativa adaptativa, se tornaram indistiguíveis por PCA (análise por componentes principais) dos espectros primários. A técnica de PDS associada a modelos PLS mostrou-se promissora, especialmente sendo capaz de estimar com boa exatidão o valor da densidade API de amostras de petróleo classificadas como medianas.

Near-infrared spectroscopy has proven to be one of the most promising techniques for the development of a noninvasive blood glucose monitoring system for diabetic patients. In this work, Fourier transform infrared (FT-IR) transmission measurements of the combination band region (4000 - 5000 cm-1) were analyzed for samples containing glucose (analyte) in a matrix of bovine serum albumin and triacetin (models for proteins and fats), all spanning physiological levels relevant for a diabetic patient. The first part of the study investigated the required spectral point-spacing for accurate detection of glucose. This was studied by systematically truncating interferograms before Fourier transforming them to single-beam spectra. A set of training data (70 samples) was collected for multivariate calibration using partial least-squares (PLS) and an external prediction set was used to verify the success of modeling glucose quantitatively. It was found that a relatively large point-spacing (16 cm-1) was successful for prediction of glucose, meaning that a shorter interferogram could be collected. The second part of the study involved collecting interferograms such that the spectral resolution was 16 cm-1, and investigating methods to extend the usefulness of calibration models for long-term data collection. Near-infrared spectroscopy often suffers from weak signals that are overwhelmed by significant instrumental drift, meaning that calibration models tend to be unsuccessful for data collected several days or months outside the calibration. For updating the calibration models, a set of 50 backgrounds containing only matrix constituents without analyte was collected on each analysis day, and used to update the original calibration model so that instrumental drift features were incorporated into the model. Background updating was found to be successful in single-beam format, producing a background-augmented (BA) PLS model that significantly improved single-beam data analysis. The standard error of prediction using the original model (PLS) and the updated model (BA-PLS) were 13.4 and 0.79 mM glucose, respectively, for a prediction set taken 176 days outside of the calibration. The matrix data also allowed for studies in background selection methods for absorbance computations as well as adaptive digital filtering that was guided by the background data.

Appropriate conservation decisions and efforts must be based on real−time and accurate information about wildlife populations. However, it is extremely challenging to monitor the population demography and physiological traits of many threatened and secretive animal species through direct observation and capture. Near infrared spectroscopy (NIRS) has the potential to be a remote tool to address questions concerning wildlife physiology and demography by analyzing “signs” of animals without seeing or capturing them. In this dissertation, two species, the giant panda (Ailuropoda melanoleuca) and red panda (Ailurus fulgens) are used as a case study, to demonstrate NIRS’ feasibility in studying their physiological properties. The aim of this study is to test NIRS’ potential as a real−time analytical tool for in the nutritional foodscape and demographic analysis using less processed or non−processed field fecal and forage samples with the help of the mode−cloning technique to transfer the master model (dry and ground samples) under laboratory conditions to satellite modes (wet or dry but unground) in field conditions. Mode−cloning is conducted using either slope and bias correction (SBC) or two spectral correction methods, piecewise direct standardization (PDS) and external parameter orthogonalization (EPO). The following four hypotheses are tested this dissertation: (1) by using mode−cloning with both SBC and PDS, unprocessed wet or unground dry bamboo leaves (pandas’ food) can be used to determine the crude protein contents; (2) machine learning−based classification models using less processed field feces after mode−cloning with spectral correction approaches (PDS and EPO) can differentiate between sexes of the giant panda; (3) mode−cloned machine learning classification models using field feces can detect pregnancy of female giant pandas; (4) with the application of mode−cloning, field fecal samples can provide sex differentiation of the red panda. This dissertation demonstrates that NIRS coupled with mode−cloning and machine learning has the potential to provide real−time and accurate prediction to determine bamboo foodscape quality and reproductive status of the giant panda and red panda using minimally processed biological samples, thus allowing quick decision-making for both in situ population monitoring of these two species and ex situ husbandry preparations for pregnant female giant pandas.

Orientador: Ronei Jesus Poppi / Tese (doutorado) - Universidade Estadual de Campinas, Instituto de Química / Made available in DSpace on 2018-08-26T13:27:08Z (GMT). No. of bitstreams: 1 Souza_AndreMarcelode_D.pdf: 34076641 bytes, checksum: 9b864adff6bf9857ffe47d6b6afee68a (MD5) Previous issue date: 2015 / Resumo: As pesquisas atuais apontam que a espectroscopia no infravermelho próximo (NIR) é a técnica alternativa mais promissora para a determinação de carbono orgânico do solo (SOC) nos laboratórios de todo o mundo em substituição total ou parcial aos métodos tradicionais de via úmida. Considerado este fato, foi desenvolvido e validado um método para a determinação de SOC por espectroscopia NIR, visando seu empregado como método de rotina em laboratórios de análise de solos do Brasil. Para este fim, foram construídos modelos de calibração multivariada a partir de um número expressivo de amostras de solos (1490 amostras, 2.980 espectros) que englobam a variabilidade de solos brasileiros. Estes modelos foram validados através da submissão dos valores previstos das concentrações de matéria orgânica do solo (SOM) ao Programa de Análise de Qualidade de Laboratórios de Fertilidade (PAQLF). As questões envolvendo a transferência de calibração entre múltiplos instrumentos também foram abordadas e a regressão por vetores de suporte (SVR) foi avaliada como alternativa à regressão em mínimos quadrados parciais (PLS). Os resultados alcançados comprovaram de maneira contundente a robustez do método proposto e indicaram que o mesmo pode substitui o método de via úmida, superando seu desempenho em alguns casos. No estudo de transferência de calibração, foi demonstrado que quando dois ou mais espectrofotômetros NIR são empregados na aquisição dos dados, recomenda-se que ambos sejam de mesma configuração. Porém, quando instrumentos diferentes foram envolvidos, o método de atualização do modelo através da matriz aumentada apresentou resultados satisfatórios em relação aos demais métodos avaliados. Existem, no entanto, pelo menos dois gargalos da implementação da espectroscopia NIR em análises de rotina: (1) o elevado custo dos instrumentos em relação ao orçamento dos laboratórios de análise de solos no Brasil; e (2) a necessidade do emprego da quimiometria na etapa de modelagem dos dados. Ambas as questões podem ser solucionadas com políticas de subsídios para compra de instrumentos e intensivos treinamentos em quimiometria e espectroscopia NIR, que podem ser oferecidos pela Empresa Brasileira de Pesquisa Agropecuária e por instituições de ensino superior do Brasil / Abstract: Current research indicates that near infrared spectroscopy (NIR) is the most promising alternative technique for the determination of soil organic carbon (SOC) in laboratories around the world in total or partial replacement traditional wet chemistry methods. Considering this fact, in this research it was developed and validated a method for determination of SOC by NIR spectroscopy, aiming its use in soils laboratories in Brazil as a routine analysis method. To this end, multivariate calibration models were constructed from a large number of soil samples (samples 1490, 2980 spectra) covering the variability of brazilian soils. These models were validated by the submission of the predicted concentrations of soil organic matter (SOM) in the Quality Program of Analysis of Fertility Laboratories (PAQLF). Issues involving the calibration transfer among several instruments were discussed and the support vector regression (SVR) was evaluated as an alternative to PLS. The results proved in a conclusive way the robustness of the proposed method and indicated that it can replace the wet chemistry method, outperforming it in some cases. Some general recommendations that can be drawn from this work are: when two or more NIR spectrophotometers are involved, it is recommended purchase or apply those of same configuration because the transfer methods evaluated generate better models when compared to those of different configuration. However, when different instruments are involved, the method of model updating through the augmented matrix showed satisfactory results when compared to other methods evaluated. There are at least two bottlenecks of the implementation of the NIR spectroscopy method in routine analysis of soil laboratories: (1) the cost of NIR spectroscopy instruments and their maintenance are considered high in relation to the budget of the laboratories of soil analysis in Brazil and (2) the spectral data treatment, which requires the use of chemometrics. Both aspects that hinder the implantation of NIR spectroscopy as a routine method can be solved, with a policy of subsidies to purchase equipment and intensive training in chemometrics that can be offered by Brazilian Corporation of Agricultural Research (Embrapa) and also by brazilian teaching institutions / Doutorado / Quimica Analitica / Doutor em Ciências

Submitted by ANA KARLA PEREIRA RODRIGUES (anakarla_@hotmail.com) on 2017-08-09T15:33:48Z No. of bitstreams: 1 arquivototal.pdf: 4699110 bytes, checksum: ef3b7c0aa5c4758d2c77e65ad6a81ad3 (MD5) / Made available in DSpace on 2017-08-09T15:33:48Z (GMT). No. of bitstreams: 1 arquivototal.pdf: 4699110 bytes, checksum: ef3b7c0aa5c4758d2c77e65ad6a81ad3 (MD5) Previous issue date: 2016-06-20 / Conselho Nacional de Pesquisa e Desenvolvimento Científico e Tecnológico - CNPq / Coordenação de Aperfeiçoamento de Pessoal de Nível Superior - CAPES / This work involves the development of two studies that are presented in chapters 2 and 3. At first, a new method to perform the calibration transfer was designed. This method was developed to make use of separate variables instead of using the full spectrum or spectral windows. To accomplish this task a univariate procedure is initially used to correct the spectra recorded in the secondary equipment, given a set of transfer samples. A robust regression technique is then used to obtain a model with small sensitivity with respect to the univariate correction. The proposed method is employed in two case studies involving near infrared spectrometric determination of specific mass, research octane number and naphtenes in gasoline, and moisture and oil in corn. In both cases, better calibration transfer results were obtained in comparison with piecewise direct standardization (PDS). In the second, a new strategy for cotton seed classification using near infrared (NIR) hyperspectral images (HSI) was developed. Initially the cotton seeds samples were recorded on a station HSI image-NIR and a conventional spectrometer NIR. Thereon, the images were segmented and the mean spectrum of each seed was extract. Classification models SPA-LDA e PLS-DA based on the mean spectral were developed for two data sets. The results for models SPA-LDA and PLSDA showed that the classification with HSI-NIR data set has been achieved with greater accuracy when compared to models for the NIR-conventional data set. / Este trabalho envolve o desenvolvimento de dois estudos, que são apresentados nos capítulos 2 e 3. No primeiro, um novo método para realizar a transferência de calibração foi concebido. Este método foi desenvolvido para fazer uso de variáveis isoladas em vez de usar todo o espectro ou janelas espectrais. Para realizar essa tarefa, um procedimento univariado é inicialmente usado para corrigir os espectros registrados no equipamento secundário, dado um conjunto de amostras de transferência. Uma técnica de regressão robusta é então usada para obter um modelo com pequena sensibilidade em relação aos resíduos da correção univariada. O novo método é então empregado em dois estudos de caso envolvendo análise espectrométrica NIR, em que foram determinados os parâmetros massa específica, RON (Research Octane Number) e teor de naftênicos em gasolina e os teores de água e óleo em amostras de milho. Os resultados do novo método foram melhores do que os obtidos usando o método PDS. No segundo, uma nova estratégia para classificação de sementes de algodão usando imagens hiperespectrais no NIR foi desenvolvido. Inicialmente as amostras de sementes de algodão foram registradas em uma estação de imagem HSI-NIR e em um equipamento NIR convencional. Após isso, as imagens foram segmentadas e os espectros médios de cada semente foram extraídos. Os modelos de classificação SPA-LDA e PLS-DA baseados nos espectros médios foram construídos para os dois conjuntos de dados. Os resultados SPA-LDA e PLS-DA para os modelos demonstraram que a classificação com os dados HSI-NIR foi alcançada com maior exatidão quando comparada aos modelos obtidos usando o NIR-convencional.

<p>  </p> <p>Hyperspectral Imaging is one of the most popular technologies in plant phenotyping due to its ability to predict the plant physiological features such as yield biomass, leaf moisture, and nitrogen content accurately, non-destructively, and efficiently. Various kinds of hyperspectral imaging systems have been developed in the past years for both greenhouse and field phenotyping activities. Developing the plant physiological prediction model such as relative water content (RWC) using hyperspectral imaging data requires the adoption of machine learning-based calibration techniques. Convolutional neural networks (CNNs) have been known to automatically extract the features from the raw data which can lead to highly accurate physiological prediction models. Once a reliable prediction model has been developed, sharing that model across multiple hyperspectral imaging systems is very desirable since collecting the large number of ground truth labels for predictive model development is expensive and tedious. However, there are always significant differences in imaging sensors, imaging, and environmental conditions between different hyperspectral imaging facilities, which makes it difficult to share plant features prediction models. Calibration transfer between the imaging systems is critically important. In this thesis, two approaches were taken to address the calibration transfer from the greenhouse to the field. First, direct standardization (DS), piecewise direct standardization (PDS), double window piecewise direct standardization (DPDS) and spectral space transfer (SST) were used for standardizing the spectral reflectance to minimize the artifacts and spectral differences between different greenhouse imaging systems. A linear transformation matrix estimated using SST based on a small set of plant samples imaged in two facilities reduced the root mean square error (RMSE) for maize physiological feature prediction significantly, i.e., from 10.64% to 2.42% for RWC and from 1.84% to 0.11% for nitrogen content. Second, common latent space features between two greenhouses or a greenhouse and field imaging system were extracted in an unsupervised fashion. Two different models based on deep adversarial domain adaptation are trained, evaluated, and tested. In contrast to linear standardization approaches developed using the same plant samples imaged in two greenhouse facilities, the domain adaptation extracted non-linear features common between spectra of different imaging systems. Results showed that transferred RWC models reduced the RMSE by up to 45.9% for the greenhouse calibration transfer and 12.4% for a greenhouse to field transfer. The plot scale evaluation of the transferred RWC model showed no significant difference between the measurements and predictions. The methods developed and reported in this study can be used to recover the performance plummeted due to the spectral differences caused by the new phenotyping system and to share the knowledge among plant phenotyping researchers and scientists.</p>