Precipitates segmentation from scanning electron microscope images through machine learning techniques

The presence of precipitates in metallic materials affects its durability, resistance and mechanical properties. Hence, its automatic identification by image processing and machine learning techniques may lead to reliable and efficient assessments on the materials. In this paper, we introduce four widely used supervised pattern recognition techniques to accomplish metallic precipitates segmentation in scanning electron microscope images from dissimilar welding on a Hastelloy C-276 alloy: Support Vector Machines, Optimum-Path Forest, Self Organizing Maps and a Bayesian classifier. Experimental results demonstrated that all classifiers achieved similar recognition rates with good results validated by an expert in metallographic image analysis. © 2011 Springer-Verlag Berlin Heidelberg.

A Markov random field model for combining optimum-path forest classifiers using decision graphs and game strategy approach

The research on multiple classifiers systems includes the creation of an ensemble of classifiers and the proper combination of the decisions. In order to combine the decisions given by classifiers, methods related to fixed rules and decision templates are often used. Therefore, the influence and relationship between classifier decisions are often not considered in the combination schemes. In this paper we propose a framework to combine classifiers using a decision graph under a random field model and a game strategy approach to obtain the final decision. The results of combining Optimum-Path Forest (OPF) classifiers using the proposed model are reported, obtaining good performance in experiments using simulated and real data sets. The results encourage the combination of OPF ensembles and the framework to design multiple classifier systems. © 2011 Springer-Verlag.

Automatic segmentation of the secondary austenite-phase island precipitates in a superduplex stainless steel weld metal

Duplex and superduplex stainless steels are class of materials of a high importance for engineering purposes, since they have good mechanical properties combination and also are very resistant to corrosion. It is known as well that the chemical composition of such steels is very important to maintain some desired properties. In the past years, some works have reported that γ 2 precipitation improves the toughness of such steels, and its quantification may reveals some important information about steel quality. Thus, we propose in this work the automatic segmentation of γ 2 precipitation using two pattern recognition techniques: Optimum-Path Forest (OPF) and a Bayesian classifier. To the best of our knowledge, this if the first time that machine learning techniques are applied into this area. The experimental results showed that both techniques achieved similar and good recognition rates. © 2012 Taylor & Francis Group.

Automatic segmentation of latent fingerprints

Latent fingerprints are routinely found at crime scenes due to the inadvertent contact of the criminals' finger tips with various objects. As such, they have been used as crucial evidence for identifying and convicting criminals by law enforcement agencies. However, compared to plain and rolled prints, latent fingerprints usually have poor quality of ridge impressions with small fingerprint area, and contain large overlap between the foreground area (friction ridge pattern) and structured or random noise in the background. Accordingly, latent fingerprint segmentation is a difficult problem. In this paper, we propose a latent fingerprint segmentation algorithm whose goal is to separate the fingerprint region (region of interest) from background. Our algorithm utilizes both ridge orientation and frequency features. The orientation tensor is used to obtain the symmetric patterns of fingerprint ridge orientation, and local Fourier analysis method is used to estimate the local ridge frequency of the latent fingerprint. Candidate fingerprint (foreground) regions are obtained for each feature type; an intersection of regions from orientation and frequency features localizes the true latent fingerprint regions. To verify the viability of the proposed segmentation algorithm, we evaluated the segmentation results in two aspects: a comparison with the ground truth foreground and matching performance based on segmented region. © 2012 IEEE.

Heterogeneous demand and market segmentation: the Argentine pension funds system

Includes bibliography

ECG arrhythmia classification based on optimum-path forest

An important tool for the heart disease diagnosis is the analysis of electrocardiogram (ECG) signals, since the non-invasive nature and simplicity of the ECG exam. According to the application, ECG data analysis consists of steps such as preprocessing, segmentation, feature extraction and classification aiming to detect cardiac arrhythmias (i.e.; cardiac rhythm abnormalities). Aiming to made a fast and accurate cardiac arrhythmia signal classification process, we apply and analyze a recent and robust supervised graph-based pattern recognition technique, the optimum-path forest (OPF) classifier. To the best of our knowledge, it is the first time that OPF classifier is used to the ECG heartbeat signal classification task. We then compare the performance (in terms of training and testing time, accuracy, specificity, and sensitivity) of the OPF classifier to the ones of other three well-known expert system classifiers, i.e.; support vector machine (SVM), Bayesian and multilayer artificial neural network (MLP), using features extracted from six main approaches considered in literature for ECG arrhythmia analysis. In our experiments, we use the MIT-BIH Arrhythmia Database and the evaluation protocol recommended by The Association for the Advancement of Medical Instrumentation. A discussion on the obtained results shows that OPF classifier presents a robust performance, i.e.; there is no need for parameter setup, as well as a high accuracy at an extremely low computational cost. Moreover, in average, the OPF classifier yielded greater performance than the MLP and SVM classifiers in terms of classification time and accuracy, and to produce quite similar performance to the Bayesian classifier, showing to be a promising technique for ECG signal analysis. © 2012 Elsevier Ltd. All rights reserved.

Unsupervised segmentation method for cuboidal cell nuclei in histological prostate images based on minimum cross entropy

This paper presents a novel segmentation method for cuboidal cell nuclei in images of prostate tissue stained with hematoxylin and eosin. The proposed method allows segmenting normal, hyperplastic and cancerous prostate images in three steps: pre-processing, segmentation of cuboidal cell nuclei and post-processing. The pre-processing step consists of applying contrast stretching to the red (R) channel to highlight the contrast of cuboidal cell nuclei. The aim of the second step is to apply global thresholding based on minimum cross entropy to generate a binary image with candidate regions for cuboidal cell nuclei. In the post-processing step, false positives are removed using the connected component method. The proposed segmentation method was applied to an image bank with 105 samples and measures of sensitivity, specificity and accuracy were compared with those provided by other segmentation approaches available in the specialized literature. The results are promising and demonstrate that the proposed method allows the segmentation of cuboidal cell nuclei with a mean accuracy of 97%. © 2013 Elsevier Ltd. All rights reserved.

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.

Motherhood wage penalties and labour market segmentation: Evidence from Argentina

Includes bibliography.

Informality and labour market segmentation: the case of Argentina

The document evaluates the presence of segmentation in the Argentinean labour market. The analysis is centred on the comparison of the earnings of formal and informal workers. Two different approaches to the definition of informality are used. The existence of a formal premium is tested using dynamic data and semiparametric techniques. The period analysed is 1996-2006 for all urban surveyed areas. Our results support the segmentation hypothesis for the Argentine urban labour market: workers with similar probabilities of entering/exiting across sectors obtain different earnings.

Análise taxonômica e molecular de Cestoda nematotaeniidae parasito de intestino delgado de Rhinella marina (Linnaeus, 1758) (Amphibia: Bufonidae) de Belém-Pa

Os anfíbios da espécie Rhinella marina também conhecidos como Sapo-Cururu e possuem distribuição mundial. Possuem hábitos noturnos, e devido a sua alimentação bem diversificada vivem em diferentes habitats. Assim podem estar parasitados com uma variedade de helmintos. Dentre os helmintos, os cestodas são o objeto de estudo deste trabalho. Os membros da Família Nematotaennidae são comumente encontrados parasitando o intestino delgado de anfíbios e répteis. O presente trabalho tem como objetivo identificar e caracterizar morfologicamente e molecularmente um cestoda parasito de R. marina da cidade de Belém-PA. Para isso vinte hospedeiros foram capturados em domicílios da região metropolitana de Belém-PA e, após necropsia, os cestoda foram retirados do intestino delgado, alguns exemplares foram fixados em A.F.A, alguns fixados em Glutaraldeído a 2% em tampão cacodilato, e outros em álcool absoluto para serem processados para diferentes técnicas. Parte da amostra foi desidratada em uma série etanólica, corados com Carmin®, clarificados com Salicilato de Metila®. Alguns exemplares foram desidratados e incluídos em parafina para realização de cortes transversais e longitudinais. Os exemplares fixados em glutaraldeído foram desidratados e incluídos em Historesina®. Os cestoda também foram processados para microscopia Eletrônica de Varredura. A identificação foi realizada por meio de desenhos realizados no microscópio Olympus BX 41 com câmara clara, fotografias feitas em microscópio MEDILUX, com sistema de captura de imagem e MEV. Os Cortes histológicos longitudinais foram fotografados e com o Software RECONSTRUCTTM foi realizada a reconstrução tridimensional do corpo do parasito. Helmintos fixados em álcool absoluto foram submetidos a extração de DNA, amplificação gênica pela técnica de PCR e seqüenciamento de nucleotídeos. Os cestoda possuem um corpo cilíndrico, filiforme e indistintamente segmentado, exceto na porção posterior. Escólice com quatro ventosas sem rostéolo ou órgão apical, os proglotes grávidos apresentam duas cápsulas piriformes, que se fundem na base, contendo os ovos. A partir das observações por microscopia eletrônica e luz dos cestoda encontrados no intestino delgado de R. marina, observou-se que estes cestoda pertencem à Família Nematotaeniidae, no entanto os outros caracteres morfológicos e moleculares por nós encontrados não encaixam este cestóide em nenhum gênero desta Família.

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.

Using Slaughter Weight to Predict Weight and Yield of Primal Cuts of Carcass from Buffaloes

The knowledge of the meat production from different buffalo breeds and their crossings in different feeding systems becomes necessary for the supply of subsidies to whole productive meat chain. Some quantitative carcass traits of Mediterranean buffaloes bulls, finished in feedlot, with initial age of fourteen months and 330 kg live weight, slaughtered with 450, 480, 510 and 540 kg, were evaluated. The diet contained 13% crude protein, 2.68 Mcal digestible energy/kg DM and a roughage : concentrate ratio of 25:75. Regression equations for prediction weight and yield of primal cuts of carcass as a function of slaughter weight were obtained. Carcass dressing percent increased as the slaughter weight increased (49.2; 49.5; 49.7; and 49.9%). The Pistola Style cut weight although increasing linearly in weight (108.2; 117.6; 124.0 and 130.7 kg) as the slaughter weight increased, declined linearly when expressed in relation to cold carcass weight (49.5; 49.0; 48.6 and 48.2%). In this experimental conditions Mediterranean young bulls slaughtered between 450 to 540 kg of live weight showed increasing yields of cold carcass, forequarter and thin flank.

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)