Aplicações da teoria de nós no estudo da estrutura de proteínas

Pós-graduação em Ciência e Tecnologia de Materiais - FC

Comparing support vector machines and artificial neural networks in the recognition of steering angle for driving of mobile robots through paths in plantations

The use of mobile robots turns out to be interesting in activities where the action of human specialist is difficult or dangerous. Mobile robots are often used for the exploration in areas of difficult access, such as rescue operations and space missions, to avoid human experts exposition to risky situations. Mobile robots are also used in agriculture for planting tasks as well as for keeping the application of pesticides within minimal amounts to mitigate environmental pollution. In this paper we present the development of a system to control the navigation of an autonomous mobile robot through tracks in plantations. Track images are used to control robot direction by pre-processing them to extract image features. Such features are then submitted to a support vector machine and an artificial neural network in order to find out the most appropriate route. A comparison of the two approaches was performed to ascertain the one presenting the best outcome. The overall goal of the project to which this work is connected is to develop a real time robot control system to be embedded into a hardware platform. In this paper we report the software implementation of a support vector machine and of an artificial neural network, which so far presented respectively around 93% and 90% accuracy in predicting the appropriate route. (C) 2013 The Authors. Published by Elsevier B.V. Selection and peer review under responsibility of the organizers of the 2013 International Conference on Computational Science

Reconhecimento de voz para aplicações em automação implementado em FPGA

In many movies of scientific fiction, machines were capable of speaking with humans. However mankind is still far away of getting those types of machines, like the famous character C3PO of Star Wars. During the last six decades the automatic speech recognition systems have been the target of many studies. Throughout these years many technics were developed to be used in applications of both software and hardware. There are many types of automatic speech recognition system, among which the one used in this work were the isolated word and independent of the speaker system, using Hidden Markov Models as the recognition system. The goals of this work is to project and synthesize the first two steps of the speech recognition system, the steps are: the speech signal acquisition and the pre-processing of the signal. Both steps were developed in a reprogrammable component named FPGA, using the VHDL hardware description language, owing to the high performance of this component and the flexibility of the language. In this work it is presented all the theory of digital signal processing, as Fast Fourier Transforms and digital filters and also all the theory of speech recognition using Hidden Markov Models and LPC processor. It is also presented all the results obtained for each one of the blocks synthesized e verified in hardware

Aplicação de técnicas de geoprocessamento para a avaliação da expansão urbana no municípo de Duque de Caxias (RJ) como subsídio ao monitoramento de dutos: trecho da faixa de dutos Rio de Janeiro - Belo Horizonte

Different forms of human pressure may occur in the pipeline ranges, due to the large extensions and various configurations of land use, which can pass through the pipelines. Due to the dynamics of these pressures, it is necessary to monitor temporal changes of land use and cover the surface. Under this theme, appears as extremely important to use products and techniques of remote sensing, as they allow the identification of objects of the land surface that may compromise the security and monitoring of the pipeline, and allows the extraction of information conditions on land use at different periods of time. Based on the above, this paper aims to examine in a temporal approach, the process of urban expansion in the municipality of Duque de Caxias, located on the outskirts of the metropolitan area of the state of Rio de Janeiro, as well as settlement patterns characteristic of areas that the changes occurred in the period 1987 to 2010. We used the technique of visual analysis to perform the change detection and the technique of image classification, aimed at monitoring human pressure over a stretch of track pipeline Rio de Janeiro - Belo Horizonte, located in the state of Rio de Janeiro. The stages of work involved the characterization of the study area, urban sprawl and the existing settlement patterns, through the analysis of bibliographic data. The processing of Landsat 5 images and the application of the technique of change detection were performed in three scenes for the years 1987, 1998 and 2010, while the classification process was performed on the image RapidEye for the year 2010. Can be noted an increase in urban area of approximately 22.38% and the change of land cover from natural to built. This growth is concentrated outside to the area of direct influence of the duct, occurring in the area of indirect influence of the enterprise. Regarding the settlement patterns of growth areas, it was observed that these are predominantly

Algorithms for automatic segmentation of bovine embryos produced

In vitro production has been employed in bovine embryos and quantification of lipids is fundamental to understand the metabolism of these embryos. This paper presents a unsupervised segmentation method for histological images of bovine embryos. In this method, the anisotropic filter was used in the differents RGB components. After pre-processing step, the thresholding technique based on maximum entropy was applied to separate lipid droplets in the histological slides in different stages: early cleavage, morula and blastocyst. In the postprocessing step, false positives are removed using the connected components technique that identify regions with excess of dye near pellucid zone. The proposed segmentation method was applied in 30 histological images of bovine embryos. Experiments were performed with the images and statistical measures of sensitivity, specificity and accuracy were calculated based on reference images (gold standard). The value of accuracy of the proposed method was 96% with standard deviation of 3%.

Estimation of croplands using indicator kriging and fuzzy classification

Coordenação de Aperfeiçoamento de Pessoal de Nível Superior (CAPES)

Classification of intact açai (Euterpe oleracea Mart.) and jucara (Euterpe edulis Mart) fruits based on dry matter content by means of near infrared spectroscopy

Fundação de Amparo à Pesquisa do Estado de São Paulo (FAPESP)

Determinação de sólidos solúveis e firmeza em frutos íntegros de pessegueiro por meio da espectroscopia do infravermelho próximo (NIR)

Coordenação de Aperfeiçoamento de Pessoal de Nível Superior (CAPES)

Gestar: formação de professores em serviço e a abordagem da geometri

This present article aims to present some results of a PhD research in which intended to answer the following investigation question: In which extent a Program of Teacher Education in service in the early years of elementary school, GESTAR, contributed to the development of positive attitudes about Geometry? The participants were twelve teachers from two schools in the State of Mato Grosso. The instruments for data collection were a range of attitudes towards Geometry (EARG) developed and experimented by Viana and Brito (2004). The results showed a meaningful change in Geometry of the data pre and post-experiment.

Um estudo sobre limites duais para o problema integrado de dimensionamento de lotes e sequenciamento da produção

Conselho Nacional de Desenvolvimento Científico e Tecnológico (CNPq)

Matrix-vector multiplication and triangular linear solver using GPGPU for symmetric positive definite matrices derived from elliptic equations

The modern GPUs are well suited for intensive computational tasks and massive parallel computation. Sparse matrix multiplication and linear triangular solver are the most important and heavily used kernels in scientific computation, and several challenges in developing a high performance kernel with the two modules is investigated. The main interest it to solve linear systems derived from the elliptic equations with triangular elements. The resulting linear system has a symmetric positive definite matrix. The sparse matrix is stored in the compressed sparse row (CSR) format. It is proposed a CUDA algorithm to execute the matrix vector multiplication using directly the CSR format. A dependence tree algorithm is used to determine which variables the linear triangular solver can determine in parallel. To increase the number of the parallel threads, a coloring graph algorithm is implemented to reorder the mesh numbering in a pre-processing phase. The proposed method is compared with parallel and serial available libraries. The results show that the proposed method improves the computation cost of the matrix vector multiplication. The pre-processing associated with the triangular solver needs to be executed just once in the proposed method. The conjugate gradient method was implemented and showed similar convergence rate for all the compared methods. The proposed method showed significant smaller execution time.

Analysis and modeling techniques for ultrasonic tissue characterization

This thesis introduces new processing techniques for computer-aided interpretation of ultrasound images with the purpose of supporting medical diagnostic. In terms of practical application, the goal of this work is the improvement of current prostate biopsy protocols by providing physicians with a visual map overlaid over ultrasound images marking regions potentially affected by disease. As far as analysis techniques are concerned, the main contributions of this work to the state-of-the-art is the introduction of deconvolution as a pre-processing step in the standard ultrasonic tissue characterization procedure to improve the diagnostic significance of ultrasonic features. This thesis also includes some innovations in ultrasound modeling, in particular the employment of a continuous-time autoregressive moving-average (CARMA) model for ultrasound signals, a new maximum-likelihood CARMA estimator based on exponential splines and the definition of CARMA parameters as new ultrasonic features able to capture scatterers concentration. Finally, concerning the clinical usefulness of the developed techniques, the main contribution of this research is showing, through a study based on medical ground truth, that a reduction in the number of sampled cores in standard prostate biopsy is possible, preserving the same diagnostic power of the current clinical protocol.

An fmri study of brain functional connectivity in parkinson's disease patients

Argomento del presente lavoro è l’analisi di dati fMRI (functional Magnetic Resonance Imaging) nell’ambito di uno studio EEG-fMRI su pazienti affetti da malattia di Parkinson idiopatica. L’EEG-fMRI combina due diverse tecniche per lo studio in vivo dell’attività cerebrale: l'elettroencefalografia (EEG) e la risonanza magnetica funzionale. La prima registra l’attività elettrica dei neuroni corticali con ottima risoluzione temporale; la seconda misura indirettamente l’attività neuronale registrando gli effetti metabolici ad essa correlati, con buona risoluzione spaziale. L’acquisizione simultanea e la combinazione dei due tipi di dati permettono di sfruttare i vantaggi di ciascuna tecnica. Scopo dello studio è l’indagine della connettività funzionale cerebrale in condizioni di riposo in pazienti con malattia di Parkinson idiopatica ad uno stadio precoce. In particolare, l’interesse è focalizzato sulle variazioni della connettività con aree motorie primarie e supplementari in seguito alla somministrazione della terapia dopaminergica. Le quattro fasi principali dell’analisi dei dati sono la correzione del rumore fisiologico, il pre-processing usuale dei dati fMRI, l’analisi di connettività “seed-based “ e la combinazione dei dati relativi ad ogni paziente in un’analisi statistica di gruppo. Usando ’elettrocardiogramma misurato contestualmente all’EEG ed una stima dell’attività respiratoria, è stata effettuata la correzione del rumore fisiologico, ottenendo risultati consistenti con la letteratura. L’analisi di connettività fMRI ha mostrato un aumento significativo della connettività dopo la somministrazione della terapia: in particolare, si è riscontrato che le aree cerebrali maggiormente connesse alle aree motorie sono quelle coinvolte nel network sensorimotorio, nel network attentivo e nel default mode network. Questi risultati suggeriscono che la terapia dopaminergica, oltre ad avere un effetto positivo sulle performance motorie durante l’esecuzione del movimento, inizia ad agire anche in condizioni di riposo, migliorando le funzioni attentive ed esecutive, componenti integranti della fase preparatoria del movimento. Nel prossimo futuro questi risultati verranno combinati con quelli ottenuti dall’analisi dei dati EEG.

The effect of diffusion gradient direction number on tractography of corticospinal tract in human brain: an along-tract analysis

Nel presente lavoro di tesi ho sviluppato un metodo di analisi di dati di DW-MRI (Diffusion-Weighted Magnetic Resonance Imaging)cerebrale, tramite un algoritmo di trattografia, per la ricostruzione del tratto corticospinale, in un campione di 25 volontari sani. Il diffusion tensor imaging (DTI) sfrutta la capacità del tensore di diffusione D di misurare il processo di diffusione dell’acqua, per stimare quantitativamente l’anisotropia dei tessuti. In particolare, nella sostanza bianca cerebrale la diffusione delle molecole di acqua è direzionata preferenzialmente lungo le fibre, mentre è ostacolata perpendicolarmente ad esse. La trattografia utilizza le informazioni ottenute tramite il DW imaging per fornire una misura della connettività strutturale fra diverse regioni del cervello. Nel lavoro si è concentrata l’attenzione sul fascio corticospinale, che è coinvolto nella motricità volontaria, trasmettendo gli impulsi dalla corteccia motoria ai motoneuroni del midollo spinale. Il lavoro si è articolato in 3 fasi. Nella prima ho sviluppato il pre-processing di immagini DW acquisite con un gradiente di diffusione sia 25 che a 64 direzioni in ognuno dei 25 volontari sani. Si è messo a punto un metodo originale ed innovativo, basato su “Regions of Interest” (ROIs), ottenute attraverso la segmentazione automatizzata della sostanza grigia e ROIs definite manualmente su un template comune a tutti i soggetti in esame. Per ricostruire il fascio si è usato un algoritmo di trattografia probabilistica che stima la direzione più probabile delle fibre e, con un numero elevato di direzioni del gradiente, riesce ad individuare, se presente, più di una direzione dominante (seconda fibra). Nella seconda parte del lavoro, ciascun fascio è stato suddiviso in 100 segmenti (percentili). Sono stati stimati anisotropia frazionaria (FA), diffusività media, probabilità di connettività, volume del fascio e della seconda fibra con un’analisi quantitativa “along-tract”, per ottenere un confronto accurato dei rispettivi percentili dei fasci nei diversi soggetti. Nella terza parte dello studio è stato fatto il confronto dei dati ottenuti a 25 e 64 direzioni del gradiente ed il confronto del fascio fra entrambi i lati. Dall’analisi statistica dei dati inter-subject e intra-subject è emersa un’elevata variabilità tra soggetti, dimostrando l’importanza di parametrizzare il tratto. I risultati ottenuti confermano che il metodo di analisi trattografica del fascio cortico-spinale messo a punto è risultato affidabile e riproducibile. Inoltre, è risultato che un’acquisizione con 25 direzioni di DTI, meglio tollerata dal paziente per la minore durata dello scan, assicura risultati attendibili. La principale applicazione clinica riguarda patologie neurodegenerative con sintomi motori sia acquisite, quali sindromi parkinsoniane sia su base genetica o la valutazione di masse endocraniche, per la definizione del grado di contiguità del fascio. Infine, sono state poste le basi per la standardizzazione dell’analisi quantitativa di altri fasci di interesse in ambito clinico o di studi di ricerca fisiopatogenetica.

Detecting Improvised Explosive Devices Via Forward Looking Ground Penetrating Radar

Forward-looking ground penetrating radar shows promise for detection of improvised explosive devices in active war zones. Because of certain insurmountable physical limitations, post-processing algorithm development is the most popular research topic in this field. One such investigative avenue explores the worthiness of frequency analysis during data post-processing. Using the finite difference time domain numerical method, simulations are run to test both mine and clutter frequency response. Mines are found to respond strongest at low frequencies and cause periodic changes in ground penetrating radar frequency results. These results are called into question, however, when clutter, a phenomenon generally known to be random, is also found to cause periodic frequency effects. Possible causes, including simulation inaccuracy, are considered. Although the clutter models used are found to be inadequately random, specular reflections of differing periodicity are found to return from both the mine and the ground. The presence of these specular reflections offers a potential alternative method of determining a mine’s presence.