The task of automatically recognizing targets in IR imagery has a history of approximately 25 years of research and development. ATR is an application of pattern recognition and scene analysis in the field of defense industry and it is still one of the challenging problems. This thesis may be viewed as an exploratory study of ATR problem with encouraging recognition algorithms implemented in the area. The examined algorithms are among the solutions to the ATR problem, which are reported to have good performance in the literature. Throughout the study, PCA, subspace LDA, ICA, nearest mean classifier, K nearest neighbors classifier, nearest neighbor classifier, LVQ classifier are implemented and their performances are compared in the aspect of recognition rate. According to the simulation results, the system, which uses the ICA as the feature extractor and LVQ as the classifier, has the best performing results. The good performance of this system is due to the higher order statistics of the data and the success of LVQ in modifying the decision boundaries.

Made available in DSpace on 2015-05-14T13:21:42Z (GMT). No. of bitstreams: 1 arquivototal.pdf: 3270377 bytes, checksum: f7faeeb5a1fdf2284d994edc54a6a265 (MD5) Previous issue date: 2013-08-26 / Coordenação de Aperfeiçoamento de Pessoal de Nível Superior - CAPES / This work presents a simple and non-expensive based on digital image and pattern recognition techniques for the classification of edible vegetable oils with respect to the type (soybean, canola, sunflower and corn) and the conservation state (expired and non-expired period of validity). For this, images of the sample oils were obtained from a webcam, and then, they were decomposed into histograms containing the distribution of color levels allowed for a pixel. Three representations for the color of a pixel were used: red-green-blue (RGB), hue-saturation-intensity (HSI) and grayscale. Linear discriminant analysis (LDA) was employed in order to build classification models on the basis of a reduced subset of variables. For the purpose of variable selection, two techniques were utilized, namely the successive projections algorithm (SPA) and stepwise (SW) formulation. Models based on partial least squares-discriminant analysis and (PLS-DA) applied to full histograms (without variable selection) were also employed for the purpose of comparison. For the study evolving the classification with respect to oil type, LDA/SPA, LDA /SW and PLS-DA models achieved a correct classification rate (CCR) of 95%, 90% and 95%, respectively. For the identification of expired non-expired samples, LDA / SPA models were found to the best method for classifying sunflower, soybean and canola oils, achieving a TCC of 97%, 94% and 93%, respectively, while the model LDA/SW correctly classified 100% of corn oil samples. These results suggest that the proposed method is a promising alternative for inspection of authenticity and the conservation state of edible vegetable oils. As advantages, the method does not use reagents to carry out the analysis and laborious procedures for chemical characterization of the samples are not required / Este trabalho apresenta um método simples e de baixo custo, baseado na utilização de imagens digitais e técnicas de reconhecimento de padrões, para a classificação de óleos vegetais comestíveis com relação ao tipo (soja, canola, girassol e milho) e ao estado de conservação (prazo de validade expirado e não expirado). Para isso, imagens das amostras de óleos vegetais foram obtidas a partir de uma webcam e, em seguida, as mesmas foram decompostas em histogramas contendo as distribuições dos níveis de cores permitidos a um pixel. Três modelos para a cor de um pixel foram utilizados: vermelho-verde-azul (RGB), matiz-saturação-intensidade (HSI) e tons de cinza. A análise discriminante linear (LDA) foi utilizada para o desenvolvimento de modelos de classificação com base em um subconjunto reduzido de variáveis. Para fins de seleção de variáveis, duas técnicas foram utilizadas: o algoritmo das projeções sucessivas (SPA) e o stepwise (SW). Modelos baseados na análise discriminante por mínimos quadrados parciais (PLS-DA) aplicados aos histogramas completos (sem seleção de variáveis) também foram utilizados com o propósito de comparação. No estudo envolvendo a classificação com respeito ao tipo, modelos LDA/SPA, LDA/SW e PLS-DA atingiram uma taxa de classificação correta (TCC) de 95%, 90% e 95%, respectivamente. Na identificação de amostras expiradas e não expiradas, o modelo LDA/SPA foi considerado o melhor método para a classificação das amostras de óleos de girassol, soja e canola, atingindo uma TCC de 97%, 94% e 93%, respectivamente, enquanto que o modelo LDA/SW classificou corretamente 100% das amostras de milho. Estes resultados sugerem que o método proposto é uma alternativa promissora para a inspeção de autenticidade e estado de conservação de óleos vegetais comestíveis. Como vantagem, a metodologia não utiliza reagentes, a análise é rápida e procedimentos laboriosos para a caracterização química das amostras não são necessários

Submitted by Luanna Matias (lua_matias@yahoo.com.br) on 2015-04-01T14:19:27Z No. of bitstreams: 2 Dissertação - Lucas de Almeida Ribeiro - 2014.pdf: 6237054 bytes, checksum: 085446421b01a7e7b9174daf3da9b192 (MD5) license_rdf: 23148 bytes, checksum: 9da0b6dfac957114c6a7714714b86306 (MD5) / Approved for entry into archive by Luanna Matias (lua_matias@yahoo.com.br) on 2015-04-01T15:19:35Z (GMT) No. of bitstreams: 2 Dissertação - Lucas de Almeida Ribeiro - 2014.pdf: 6237054 bytes, checksum: 085446421b01a7e7b9174daf3da9b192 (MD5) license_rdf: 23148 bytes, checksum: 9da0b6dfac957114c6a7714714b86306 (MD5) / Made available in DSpace on 2015-04-01T15:19:35Z (GMT). No. of bitstreams: 2 Dissertação - Lucas de Almeida Ribeiro - 2014.pdf: 6237054 bytes, checksum: 085446421b01a7e7b9174daf3da9b192 (MD5) license_rdf: 23148 bytes, checksum: 9da0b6dfac957114c6a7714714b86306 (MD5) Previous issue date: 2014-10-29 / Coordenação de Aperfeiçoamento de Pessoal de Nível Superior - CAPES / This work proposes the use of multi-objective evolutionary algorithm on tables (AEMT) for variable selection in classification problems, using linear discriminant analysis. The proposed algorithm aims to find minimal subsets of the original variables, robust classifiers that model without significant loss in classification ability. The results of the classifiers modeled by the solutions found by this algorithm are compared in this work to those found by mono-objective formulations (such as PLS, APS and own implementations of a Simple Genetic Algorithm) and multi-objective formulations (such as the simple genetic algorithm multi -objective - MULTI-GA - and the NSGA II). As a case study, the algorithm was applied in the selection of spectral variables for classification by linear discriminant analysis (LDA) of samples of biodiesel / diesel. The results showed that the evolutionary formulations are solutions with a smaller number of variables (on average) and a better error rate (average) and compared to the PLS APS. The formulation of the AEMT proposal with the fitness functions: medium risk classification, number of selected variables and number of correlated variables in the model, found solutions with a lower average errors found by the NSGA II and the MULTI-GA, and also a smaller number of variables compared to the multi-GA. Regarding the sensitivity to noise the solution found by AEMT was less sensitive than other formulations compared, showing that the AEMT is more robust classifiers. Finally shows the separation regions of classes, based on the dispersion of samples, depending on the selected one of the solutions AEMT, it is noted that it is possible to determine variables of regions split from the selected variables. / Este trabalho propõe o uso do algoritmo evolutivo multi-objetivo em tabelas (AEMT) para a seleção de variáveis em problemas de classificação, por meio de análise discriminante linear. O algoritmo proposto busca encontrar subconjuntos mínimos, das variáveis originais, que modelem classificadores robustos, sem perda significativa na capacidade de classificação. Os resultados dos classificadores modelados pelas soluções encontradas por este algoritmo são comparadas, neste trabalho, às encontradas por formulações mono-objetivo (como o PLS, o APS e implementações próprias de um Algoritmo Genético Simples) e formulações multi-objetivos (como algoritmo genético multi-objetivo simples - MULTI-GA - e o NSGA II). Como estudo de caso, o algoritmo foi aplicado na seleção de variáveis espectrais, para a classificação por análise discriminante linear (LDA - Linear Discriminant Analysis), de amostras de biodiesel/diesel. Os resultados obtidos mostraram que as formulações evolutivas encontram soluções com um menor número de variáveis (em média) e uma melhor taxa de erros (média) se comparadas ao PLS e o APS. A formulação do AEMT proposta com as funções de aptidão: risco médio de classificação, número de variáveis selecionadas e quantidade de variáveis correlacionadas presentes no modelo, encontrou soluções com uma média de erros inferior as encontradas pelo NSGA II e pelo MULTI-GA, e também uma menor quantidade de variáveis se comparado ao MULTI-GA. Em relação à sensibilidade a ruídos a solução encontrada pelo AEMT se mostrou menos sensível que as outras formulações comparadas, mostrando assim que o AEMT encontra classificadores mais robustos. Por fim, são apresentadas as regiões de separação das classes, com base na dispersão das amostras, em função das variáveis selecionadas por uma das soluções do AEMT, nota-se que é possível determinar regiões de separação a partir das variáveis selecionadas.

Context. Human oracle cost, the cost associated in estimating the correctness of the output for the given test inputs is manually evaluated by humans and this cost is significant and is a concern in the software test data generation field. This study has been designed in the context to assess metrics that might predict human oracle cost. Objectives. The major objective of this study is to address the human oracle cost, for this the study identifies the metrics that are good predictors of human oracle cost and can further help to solve the oracle problem. In this process, the identified suitable metrics from the literature are applied on the test input, to see if they can help in predicting the correctness of the output for the given test input. Methods. Initially a literature review was conducted to find some of the metrics that are relevant to the test data. Besides finding the aforementioned metrics, our literature review also tries to find out some possible code metrics that can be ap- plied on test data. Before conducting the actual experiment two pilot experiments were conducted. To accomplish our research objectives an experiment is conducted in the BTH university with master students as sample population. Further group interviews were conducted to check if the participants perceive any new metrics that might impact the correctness of the output. The data obtained from the experiment and the interviews is analyzed using linear regression model in SPSS suite. Further to analyze the accuracy vs metric data, linear discriminant model using SPSS pro- gram suite was used. Results.Our literature review resulted in 4 metrics that are suitable to our study. As our test input is HTML we took HTML depth, size, compression size, number of tags as our metrics. Also, from the group interviews another 4 metrics are drawn namely number of lines of code and number of <div>, anchor <a> and paragraph <p> tags as each individual metric. The linear regression model which analyses time vs metric data, shows significant results, but with multicollinearity effecting the result, there was no variance among the considered metrics. So, the results of our study are proposed by adjusting the multicollinearity. Besides, the above analysis, linear discriminant model which analyses accuracy vs metric data was conducted to predict the metrics that influences accuracy. The results of our study show that metrics positively correlate with time and accuracy. Conclusions. From the time vs metric data, when multicollinearity is adjusted by applying step-wise regression reduction technique, the program size, compression size and <div> tag are influencing the time taken by sample population. From accuracy vs metrics data number of <div> tags and number of lines of code are influencing the accuracy of the sample population.

New experimental strategies and algorithmic approaches were devised and tested to improve the analysis of pharmaceutically relevant materials. These new methods were developed to address key bottlenecks in the design of amorphous solid dispersions for the delivery of low-solubility active pharmaceutical ingredients in the final dosage forms exhibiting high bioavailability. <br>

In this paper, we propose techniques for adaptation of speaker recognition systems. The aim of this work is to create adaptation for Probabilistic Linear Discriminant Analysis. Special attention is given to unsupervised adaptation. Our test shows appropriate clustering techniques for speaker estimation of the identity and estimation of the number of speakers in adaptation dataset. For the test, we are using NIST and Switchboard corpora.

This document introduces the reader to area of face recognition. Miscellaneous methods are mentioned and categorized to be able to understand the process of face recognition. Main focus of this document is on issues of current face recognition and possibilities do solve these inconveniences in order to be able to massively spread face recognition. The second part of this work is focused on implementation of selected methods, which are Linear Discriminant Analysis and Principal Component Analysis. Those methods are compared to each other and results are given at the end of work.

Indiana University-Purdue University Indianapolis (IUPUI) / The link between volatile organic compounds (VOCs) from breath and various diseases and specific conditions has been identified since long by the researchers. Canine studies and breath sample analysis on Gas chromatography/ Mass Spectroscopy has proven that there are VOCs in the breath that can detect and potentially predict hypoglycemia. This project aims at developing a smart nanosensor system to detect hypoglycemia from human breath. The sensor system comprises of 1-Mercapto-(triethylene glycol) methyl ether functionalized goldnanoparticle (EGNPs) sensors coated with polyetherimide (PEI) and poly(vinylidene fluoride -hexafluoropropylene) (PVDF-HFP) and polymer composite sensor made from PVDF-HFP-Carbon Black (PVDF-HFP/CB), an interface circuit that performs signal conditioning and amplification, and a microcontroller with Bluetooth Low Energy (BLE) to control the interface circuit and communicate with an external personal digital assistant. The sensors were fabricated and tested with 5 VOCs in dry air and simulated breath (a mixture of air, small portion of acetone, ethanol at high humidity) to investigate sensitivity and selectivity. The name of the VOCs is not disclosed herein but these VOCs have been identified in-breath and are identified as potential biomarkers for other diseases as well. The sensor hydrophobicity has been studied using contact angle measurement. The GNPs size was verified using Ultra-Violent-Visible (UV-VIS) Spectroscopy. Field Emission Scanning Electron Microscope (FESEM) image is used to show GNPs embedded in the polymer film. The sensors sensitivity increases by more than 400\% in an environment with relative humidity (RH) of 93\% and the sensors show selectivity towards VOCs of interest. The interface circuit was designed on Eagle PCB and was fabricated using a two-layer PCB. The fabricated interface circuit was simulated with variable resistance and was verified with experiments. The system is also tested at different power source voltages and it was found that the system performance is optimum at more than 5 volts. The sensor fabrication, testing methods, and results are presented and discussed along with interface circuit design, fabrication, and characterization. / 2022-05-8

Betygsättning upptar idag en stor del av lärares arbetstid och det finns en betydande inkonsekvens vid bedömning utförd av olika lärare. Denna studie ämnar undersöka vilken träffsäkerhet som en automtiserad bedömningsmodell kan uppnå. Tre maskininlärningsmodeller för klassifikation i form av Linear Discriminant Analysis, K-Nearest Neighbor och Random Forest tränas och testas med femfaldig korsvalidering på uppsatser från nationella prov i svenska. Klassificeringen baseras på språk och formrelaterade attribut inkluderande ord och teckenvisa längdmått, likhet med texter av olika formalitetsgrad och grammatikrelaterade mått. Detta utmynnar i ett maximalt quadratic weighted kappa-värde på 0,4829 och identisk överensstämmelse med expertgivna betyg i 57,53 % av fallen. Dessa resultat uppnåddes av en modell baserad på Linear Discriminant Analysis och uppvisar en högre korrelation med expertgivna betyg än en ordinarie lärare. Trots pågående digitalisering inom skolväsendet kvarstår ett antal hinder innan fullständigt maskininlärningsbaserad bedömning kan realiseras, såsom användarnas inställning till tekniken, etiska dilemman och teknikens svårigheter med förståelse av semantik. En delvis integrerad automatisk betygssättning har dock potential att identifiera uppsatser där behov av dubbelrättning föreligger, vilket kan öka överensstämmelsen vid storskaliga prov till en låg kostnad. / Today, a large amount of a teacher’s workload is comprised of essay scoring and there is a large variability between teachers’ gradings. This report aims to examine what accuracy can be acceived with an automated essay scoring system for Swedish. Three following machine learning models for classification are trained and tested with 5-fold cross-validation on essays from Swedish national tests: Linear Discriminant Analysis, K-Nearest Neighbour and Random Forest. Essays are classified based on 31 language structure related attributes such as token-based length measures, similarity to texts with different formal levels and use of grammar. The results show a maximal quadratic weighted kappa value of 0.4829 and a grading identical to expert’s assessment in 57.53% of all tests. These results were achieved by a model based on Linear Discriminant Analysis and showed higher inter-rater reliability with expert grading than a local teacher. Despite an ongoing digitilization within the Swedish educational system, there are a number of obstacles preventing a complete automization of essay scoring such as users’ attitude, ethical issues and the current techniques difficulties in understanding semantics. Nevertheless, a partial integration of automatic essay scoring has potential to effectively identify essays suitable for double grading which can increase the consistency of large-scale tests to a low cost.

Classification of data from repeated measurements is useful in various disciplines, for example that of medicine. This thesis explores how classification trees (CART) can be used for classifying repeated measures data. The reader is introduced to variations of the CART algorithm which can be used for classifying the data set and tests the performance of these algorithms on a data set that can be modelled using bilinear regression. The performance is compared with that of a classification rule based on linear discriminant analysis. It is found that while the performance of the CART algorithm can be satisfactory, using linear discriminant analysis is more reliable for achieving good results. / Klassificering av data från upprepade mätningar är användbart inom olika discipliner, till exempel medicin. Denna uppsats undersöker hur klassificeringsträd (CART) kan användas för att klassificera upprepade mätningar. Läsaren introduceras till varianter av CART-algoritmen som kan användas för att klassificera datamängden och testar prestandan för dessa algoritmer på en datamängd som kan modelleras med hjälp av bilinjär regression. Prestandan jämförs med en klassificeringsregel baserad på linjär diskriminantanalys. Det har visar sig att även om prestandan för CART-algoritmen kan vara tillfredsställande, är användning av linjär diskriminantanalys mer tillförlitlig för att uppnå goda resultat.