Contribuições para o Atlas Prosódico Multimídia do Português do Norte do Brasil – AMPER-POR: variedade linguística do município de Abaetetuba (PA)

O objetivo principal dessa Dissertação de Mestrado é caracterizar a variação prosódica dialetal do português falado no município de Abaetetuba (PA). Todos os procedimentos metodológicos adotados, aqui, neste estudo, seguem as orientações estabelecidas pela equipe do Projeto AMPER, na condução do tratamento dos dados, para a confecção do Atlas Prosódico Multimídia das Línguas Românicas. As produções linguísticas dos falantes foram gravadas usando um único padrão, garantindo uma produção do sinal acústico de qualidade uniforme e uma boa representatividade da variedade dialetal. O corpus é constituído de 102 frases, SVC (sujeito + verbo + complemento) e suas expansões (sintagma adjetival e preposicionado), estruturadas com as mesmas restrições fonéticas e sintáticas. Cada uma das sentenças foi repetida seis vezes, por cada um dos quatro informantes, e o corpus total é composto por 612 frases. O pitch, para os informantes do sexo masculino, está entre 50 Hz e 250 Hz; e 110 Hz a 370 Hz para os informantes do sexo feminino. Foram utilizados três parâmetros acústicos controlados: a Frequência fundamental (F0), a Duração (ms) e a Intensidade (dB). O tratamento dos dados foi realizado por meio de sete etapas: 1) codificação das repetições, 2) isolamento de cada sentença em áudio individual; 3) segmentação fonética realizado no software PRAAT; 4) aplicação do PRAAT script; 5) seleção das três melhores repetições; 6) aplicação da interface MATLAB; e 7) utilização do EXCEL para gerar os gráficos para análise comparativa dos dados. Os resultados mostram que “as três maiores variações dos parâmetros acústicos controlados ocorrem preferencialmente na sílaba tônica da parte central do sintagma e/ou no sintagma final do enunciado” (CRUZ; BRITO, 2011).

Estudo perceptual da prosódia como elemento delimitador da estrutura de narrativas orais espontâneas: a diferença de tom

A presente pesquisa tem como tema o estudo perceptual da prosódia como elemento de segmentação de narrativas orais espontâneas e visa confirmar, ou não, se a prosódia facilita ao ouvinte leigo e inexperiente perceber a estrutura do texto narrativo. Este estudo investiga se a diferença de tom é um elemento prosódico relevante. A dissertação tem como corpus quatro narrativas espontâneas, as quais fazem parte do corpus analisado por Oliveira Jr.(2000), autor do projeto que inspirou esta pesquisa. Para saber se os participantes são capazes de delimitar a estrutura narrativa, baseando-se apenas no aspecto perceptual, conduziu-se um teste de percepção com 112 voluntários, recrutados na Universidade Federal do Pará e na Universidade Federal de Alagoas. Coube aos participantes a tarefa de indicar os pontos em que o falante teve a intenção de finalizar uma unidade comunicativa nas narrativas. A interpretação sobre unidade comunicativa foi subjetiva. Apresentou-se cada narrativa em quatro condições diferentes, a saber: (i) transcrição sem marca de pontuação e sem paragrafação; (ii) transcrição da narrativa acompanhada de áudio ; (iii) narrativa somente em áudio e (iv) áudio filtrado da narrativa, resultando numa versão deslexicalizada (fala ininteligível), mas com preservação da estrutura prosódica do discurso. Nas duas primeiras condições, a segmentação foi no texto transcrito, com barras transversais (/); nas demais, utilizou-se um programa de computador chamado ELAN. A análise dos dados obtidos baseou-se em tabelas, gráficos, análise estatística (teste do Qui-Quadrado), análise acústica (utilização do Programa PRAAT). Os resultados sinalizam que a prosódia ajuda o ouvinte leigo a perceber a estrutura básica do discurso narrativo. Com relação ao peso do Pitch Reset para auxiliar os ouvintes na demarcação de fronteiras, pode-se dizer que o teste estatístico do Qui-Quadrado encontrou evidências que lhe atribui essa função. Assim, neste contexto, ratifica-se o relevante papel da prosódia para o reconhecimento da estrutura de narrativas orais espontâneas e identifica-se o reflexo do peso da diferença de tom na percepção dos participantes.

Application of two segmentation protocols during the processing of virtual images in rapid prototyping: ex vivo study with human dry mandibles

The aim of this study was to evaluate the accuracy of virtual three-dimensional (3D) reconstructions of human dry mandibles, produced from two segmentation protocols (outline only and all-boundary lines).Twenty virtual three-dimensional (3D) images were built from computed tomography exam (CT) of 10 dry mandibles, in which linear measurements between anatomical landmarks were obtained and compared to an error probability of 5 %.The results showed no statistically significant difference among the dry mandibles and the virtual 3D reconstructions produced from segmentation protocols tested (p = 0,24).During the designing of a virtual 3D reconstruction, both outline only and all-boundary lines segmentation protocols can be used.Virtual processing of CT images is the most complex stage during the manufacture of the biomodel. Establishing a better protocol during this phase allows the construction of a biomodel with characteristics that are closer to the original anatomical structures. This is essential to ensure a correct preoperative planning and a suitable treatment.

Comparative analysis of two techniques for detection and segmentation of moving bodies

This paper makes a comparative analysis of results produced by the application of two techniques for the detection and segmentation of bodies in motion captured in images sequence, namely: 1) technique based on the temporal average of the values of each pixel recorded in N consecutive image frames and, 2) technique based on historical values associated with pixels recorded in different frames of an image sequence.

An automatic method for segmentation of fission tracks in epidote crystal photomicrographs

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

Segmentation of Scanning Electron Microscopy Images From Natural Rubber Samples With Gold Nanoparticles Using Starlet Wavelets

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

Proposta teórico-metodológica para análise de inserções parentéticas em chat educacional no ensino de língua espanhola =: Propuesta teórico-metodológica para el análisis de inserciones parentéticas en chat educacional en la enseñanza de lengua española

Conselho Nacional de Desenvolvimento Científico e Tecnológico (CNPq)

Estimating the concentration of gold nanoparticles incorporated on natural rubber membranes using multi-level starlet optimal segmentation

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

Produção e utilização de material didático sobre fisiologia da audição e neurofisiologia

The teaching of hearing physiology requires an knowledge integration of Human Anatomy, Biophysics, more precisely Bioacoustics and Bioelectrogenesis, as well as Neurophysiology. Students present difficulty to build knowledge about functional mechanisms of sound conduction and sensory transduction, especially if the elements are not visible forms, as the middle and inner ear structures. To make the teaching about hearing physiology and sensory perception easier, was produced a set of didactical materials about the subject. At first, a resin model that faithfully describes the anatomical relationship of the ossicles with the tympanic membrane was developed. Subsequently, a second model that, besides illustrates the mechanism of acoustic impedance overcoming, also reveals how acoustic sensorial transduction occurs in inner ear, was designed and produced. In the third didactical model, are visualized, through students interaction, areas of the cerebral cortex that interpret the different sensory modalities. In addition, were created three educational videos about hearing problems and a site on Human Hearing Physiology, available on Institute of Biosciences website. The results of this course conclusion monograph are presented in the form of articles that were submitted to Journal Physics in the School and the Journal of the Nucleus of Teaching

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%.

Segmentation of Cell Nuclei Regions in Epithelium of Prostate Glands

This paper proposes a method for segmentation of cell nuclei regions in epithelium of prostate glands. This structure provides information to diagnosis and prognosis of prostate cancer. In the initial step, the contrast stretching technique was applied in image in order to improve the contrast between regions of interest and other regions. After, the global thresholding technique was applied and the value of threshold was defined empirically. Finally, the false positive regions were removed using the connected components technique. The performance of the proposed method was compared with the Otsu technique and statistical measures of accuracy were calculated based on reference images (gold standard). The result of the mean value of accuracy of proposed method was 93% ± 0.07.

Image segmentation through combined methods: watershed transform, unsupervised distance learning and normalized cut

Research on image processing has shown that combining segmentation methods may lead to a solid approach to extract semantic information from different sort of images. Within this context, the Normalized Cut (NCut) is usually used as a final partitioning tool for graphs modeled in some chosen method. This work explores the Watershed Transform as a modeling tool, using different criteria of the hierarchical Watershed to convert an image into an adjacency graph. The Watershed is combined with an unsupervised distance learning step that redistributes the graph weights and redefines the Similarity matrix, before the final segmentation step using NCut. Adopting the Berkeley Segmentation Data Set and Benchmark as a background, our goal is to compare the results obtained for this method with previous work to validate its performance.

Image segmentation using morphological watershed applied to cartography

Image segmentation is a process frequently used in several different areas including Cartography. Feature extraction is a very troublesome task, and successful results require more complex techniques and good quality data. The aims of this paper is to study Digital Image Processing techniques, with emphasis in Mathematical Morphology, to use Remote Sensing imagery, making image segmentation, using morphological operators, mainly the multi-scale morphological gradient operator. In the segmentation process, pre-processing operators of Mathematical Morphology were used, and the multi-scales gradient was implemented to create one of the images used as marker image. Orbital image of the Landsat satellite, sensor TM was used. The MATLAB software was used in the implementation of the routines. With the accomplishment of tests, the performance of the implemented operators was verified and carried through the analysis of the results. The extration of linear feature, using mathematical morphology techniques, can contribute in cartographic applications, as cartographic products updating. The comparison to the best result obtained was performed by means of the morphology with conventional techniques of features extraction. © Springer-Verlag 2004.

Sensibilização da audição e educação sonora nos exercícios de limpeza de ouvidos: uma incursão na Educação Matemática

The teaching of acoustics has been characterized by a banking model that little contributes to the ideal training of citizens capable of understanding and acting to improve their environment soundscapes. Equally distant from the world of sound and musical culture, audio technology and acoustic environment, it is disconnected from the ever-increasing effort to raise awareness on hearing and sound education, as defended by the Canadian educator Prof. Raymond Murray Schafer. In order to provide elements for reflection on how Mathematics can be itself a language to compete in a sound education, we developed, in a dialogical and problematizing method applied to the technological and cultural world, one further research and teaching with Math students of UNEMAT in Barra do Bugres. This study pointed to the feasibility of educating consciences capable of of improving their acoustic environment, modifying the landscapes where we live, under our responsibility.