Robust and fast vowel recognition using optimum-path forest

The applications of Automatic Vowel Recognition (AVR), which is a sub-part of fundamental importance in most of the speech processing systems, vary from automatic interpretation of spoken language to biometrics. State-of-the-art systems for AVR are based on traditional machine learning models such as Artificial Neural Networks (ANNs) and Support Vector Machines (SVMs), however, such classifiers can not deal with efficiency and effectiveness at the same time, existing a gap to be explored when real-time processing is required. In this work, we present an algorithm for AVR based on the Optimum-Path Forest (OPF), which is an emergent pattern recognition technique recently introduced in literature. Adopting a supervised training procedure and using speech tags from two public datasets, we observed that OPF has outperformed ANNs, SVMs, plus other classifiers, in terms of training time and accuracy. ©2010 IEEE.

O teatro experimental do negro -: estudo da personagem negra em duas peças encenadas (1947-1951)

Pós-graduação em Artes - IA

Partial HELLP Syndrome: maternal and perinatal outcome

CONTEXTO: A síndrome HELLP é uma grave complicação da gestação caracterizada por hemólise, elevação das enzimas hepáticas e plaquetopenia. Algumas gestantes desenvolvem somente uma ou duas dessas características da síndrome HELLP. Esse quadro é denominado de síndrome HELLP parcial (SHP). OBJETIVO: O objetivo deste estudo foi avaliar as repercussões maternas e perinatais das mulheres que desenvolveram SHP e comparar os resultados com mulheres que tiveram hipertensão gestacional ou pré-eclâmpsia sem alterações dos exames laboratoriais para síndrome HELLP. TIPO DE ESTUDO: Observacional, retrospectivo e analítico. LOCAL: Maternidade do Hospital das Clínicas da Faculdade de Medicina de Botucatu, Universidade Estadual Paulista, Botucatu, São Paulo, Brasil. AMOSTRA: Foram selecionadas gestantes ou puérperas que tiveram elevação dos níveis pressóricos detectada pela primeira vez após a primeira metade da gestação com ou sem proteinúria entre janeiro/1990 a dezembro/1995. As mulheres foram divididas em dois grupos: Grupo SHP quando as mulheres com hipertensão arterial tinham pelo menos uma, mas não todas as alterações de exames que demonstravam hemólise, elevação das enzimas hepáticas ou plaquetopenia e Grupo Hipertensas pacientes com hipertensão sem alterações nos exames laboratoriais para síndrome HELLP. PRINCIPAIS VARIÁVEIS: Analisamos idade materna, raça, paridade, classificação da hipertensão, idade gestacional no diagnóstico da SHP, alterações nos exames laboratoriais para síndrome HELLP, tempo de permanência no hospital, complicações maternas, via de parto, incidência de prematuridade, restrição de crescimento intra-uterino, natimortos e neomortos. RESULTADOS: 318 mulheres foram selecionadas, das quais 41 (12,9%) tiveram SHP e 277 (87,1%) não desenvolveram alterações dos exames laboratoriais que compõem o diagnóstico da síndrome HELLP. A pré-eclâmpsia foi um tipo de hipertensão mais freqüente no grupo SHP que no grupo hipertensas. Não houve pacientes com hipertensão crônica isolada que desenvolveram SHP. A taxa de cesárea, eclâmpsia e de partos prematuros foi significativamente mais freqüente no grupo SHP que no grupo hipertensas. CONCLUSÃO: Observamos uma conduta agressiva nas pacientes com SHP, que resultou na interrupção imediata da gestação, com elevada taxa de cesárea e de recém-nascido pré-termo. Esta conduta deve ser revista para a redução desses índices.

Feature selection through gravitational search algorithm

In this paper we deal with the problem of feature selection by introducing a new approach based on Gravitational Search Algorithm (GSA). The proposed algorithm combines the optimization behavior of GSA together with the speed of Optimum-Path Forest (OPF) classifier in order to provide a fast and accurate framework for feature selection. Experiments on datasets obtained from a wide range of applications, such as vowel recognition, image classification and fraud detection in power distribution systems are conducted in order to asses the robustness of the proposed technique against Principal Component Analysis (PCA), Linear Discriminant Analysis (LDA) and a Particle Swarm Optimization (PSO)-based algorithm for feature selection.

Optimum-Path Forest pruning parameter estimation through Harmony Search

Pattern recognition in large amount of data has been paramount in the last decade, since that is not straightforward to design interactive and real time classification systems. Very recently, the Optimum-Path Forest classifier was proposed to overcome such limitations, together with its training set pruning algorithm, which requires a parameter that has been empirically set up to date. In this paper, we propose a Harmony Search-based algorithm that can find near optimal values for that. The experimental results have showed that our algorithm is able to find proper values for the OPF pruning algorithm parameter. © 2011 IEEE.

Análise e extração de características de imagens termográficas utilizando componentes principais

Pós-graduação em Engenharia Mecânica - FEIS

Avaliação do significado social, econômico e cultural do tabagismo em mulheres profissionais do sexo

Pós-graduação em Enfermagem - FMB

An ontology based for drilling report classification

This paper presents an application of an ontology based system for automated text analysis using a sample of a drilling report to demonstrate how the methodology works. The methodology used here consists basically of organizing the knowledge related to the drilling process by elaborating the ontology of some typical problems. The whole process was carried out with the assistance of a drilling expert, and by also using software to collect the knowledge from the texts. Finally, a sample of drilling reports was used to test the system, evaluating its performance on automated text classification.

Automatic segmentation and classification of human intestinal parasites from microscopy images

Human intestinal parasites constitute a problem in most tropical countries, causing death or physical and mental disorders. Their diagnosis usually relies on the visual analysis of microscopy images, with error rates that may range from moderate to high. The problem has been addressed via computational image analysis, but only for a few species and images free of fecal impurities. In routine, fecal impurities are a real challenge for automatic image analysis. We have circumvented this problem by a method that can segment and classify, from bright field microscopy images with fecal impurities, the 15 most common species of protozoan cysts, helminth eggs, and larvae in Brazil. Our approach exploits ellipse matching and image foresting transform for image segmentation, multiple object descriptors and their optimum combination by genetic programming for object representation, and the optimum-path forest classifier for object recognition. The results indicate that our method is a promising approach toward the fully automation of the enteroparasitosis diagnosis. © 2012 IEEE.

Aquatic weed automatic classification using machine learning techniques

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

Vibrational spectroscopy applied to the study of archeological ceramic artifacts from Guarani culture in Brazil

An archeological artifact can be seen as a chronological element, which helps to determine the age of certain society and to understand the thinking, values and the way of life of this society. Thus, the classification of archeological artifacts is one of the approaches used to study the cultural system of antique societies trying to reconstruct their history. The "Centro de Museologia, Antropologia e Argueologia (CEMAARQ)" of the "Unesp Univ Estadual Paulista" in Presidente Prudente, São Paulo state, Brazil, develops projects within this context (identification and preservation). This is the case of the archeological site named "Lagoa São Paulo-02" discovered in 1993 at the margins of the Parana river in the region of Presidente Epitacio city, São Paulo state, Brazil. This site has ceramic fragments of different shapes and sizes that have a strong influence of traces of the Guarani culture, which is one of the Brazilian native populations. These samples were basically characterized via micro-Raman scattering and Fourier transform infrared absorption (FTIR) spectroscopies. The main objective was to identify the pigments used in the manufacture of the ceramic artifacts and to analyze the composition of the ceramic body to understand how the artifacts were made. Three pigments were found: red, black and white. For the red pigment were identified characteristic bands of hematite, an iron oxide found in the red rocks of the river banks that were eroded by water. The black pigment, probably, is due to the use of vegetal charcoal, which is found in nature as the product of burning organic material such as wood. For the white pigment, the FTIR spectra suggested the use of kaolin, either in the ceramic body or in the proper white pigment, due to the presence of the characteristic bands of the kaolinite. Complementary, the additives applied as anti-plastics were identified as charcoal and quartz, being the latter found in the rocks present in the archeological site. (C) 2010 Elsevier B.V. All rights reserved.

Efficient supervised optimum-path forest classification for large datasets

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

CENSUS HISTOGRAMS: A SIMPLE FEATURE EXTRACTION and MATCHING APPROACH FOR FACE RECOGNITION

Most face recognition approaches require a prior training where a given distribution of faces is assumed to further predict the identity of test faces. Such an approach may experience difficulty in identifying faces belonging to distributions different from the one provided during the training. A face recognition technique that performs well regardless of training is, therefore, interesting to consider as a basis of more sophisticated methods. In this work, the Census Transform is applied to describe the faces. Based on a scanning window which extracts local histograms of Census Features, we present a method that directly matches face samples. With this simple technique, 97.2% of the faces in the FERET fa/fb test were correctly recognized. Despite being an easy test set, we have found no other approaches in literature regarding straight comparisons of faces with such a performance. Also, a window for further improvement is presented. Among other techniques, we demonstrate how the use of SVMs over the Census Histogram representation can increase the recognition performance.

Evidence supporting the recognition of Cole latent virus as a distinct carlavirus

Some molecular properties are described of Cole latent virus (CoLV), hitherto designated a tentative species of the Carlavirus genus. CoLV genomic RNA (Ribonucleic acid) of 8.3 Kb is polyadenylated. Two unencapsidated polyadenylated subgenomic RNAs (2.6 and 1.3 Kb) and three double-stranded RNAs (dsRNAs) (8.3, 2.6 and 1.3 Kbp), which are twice the size of the genomic and subgenomics ssRNAs, are produced in CoLV-infected plants, two additional dsRNAs (7.2 and 6.3 Kbp) were also detected plant extracts. By using a Carlavirus specific primer and a CoLV cDNA, a-3'-terminus fragment of 116 bp was amplified; it had homology with the carlaviruses Potato virus M (62%)., Hop latent virus (37%) and Blueberry scorch virus (36%) but no significant homology with 11 other carlaviruses. These results support the classification of CoLV as a distinct species of the Carlavirus genus.

Artificial immune systems for classification of petroleum well drilling operations

This paper presents two approaches of Artificial Immune System for Pattern Recognition (CLONALG and Parallel AIRS2) to classify automatically the well drilling operation stages. The classification is carried out through the analysis of some mud-logging parameters. In order to validate the performance of AIS techniques, the results were compared with others classification methods: neural network, support vector machine and lazy learning.