Optimization of neural classifiers based on bayesian decision boundaries and idle neurons pruning

In this article we describe a feature extraction algorithm for pattern classification based on Bayesian Decision Boundaries and Pruning techniques. The proposed method is capable of optimizing MLP neural classifiers by retaining those neurons in the hidden layer that realy contribute to correct classification. Also in this article we proposed a method which defines a plausible number of neurons in the hidden layer based on the stem-and-leaf graphics of training samples. Experimental investigation reveals the efficiency of the proposed method. © 2002 IEEE.

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.

Applying machine learning based on multiscale classifiers to detect remote phenology patterns in Cerrado savanna trees

Plant phenology is one of the most reliable indicators of species responses to global climate change, motivating the development of new technologies for phenological monitoring. Digital cameras or near remote systems have been efficiently applied as multi-channel imaging sensors, where leaf color information is extracted from the RGB (Red, Green, and Blue) color channels, and the changes in green levels are used to infer leafing patterns of plant species. In this scenario, texture information is a great ally for image analysis that has been little used in phenology studies. We monitored leaf-changing patterns of Cerrado savanna vegetation by taking daily digital images. We extract RGB channels from the digital images and correlate them with phenological changes. Additionally, we benefit from the inclusion of textural metrics for quantifying spatial heterogeneity. Our first goals are: (1) to test if color change information is able to characterize the phenological pattern of a group of species; (2) to test if the temporal variation in image texture is useful to distinguish plant species; and (3) to test if individuals from the same species may be automatically identified using digital images. In this paper, we present a machine learning approach based on multiscale classifiers to detect phenological patterns in the digital images. Our results indicate that: (1) extreme hours (morning and afternoon) are the best for identifying plant species; (2) different plant species present a different behavior with respect to the color change information; and (3) texture variation along temporal images is promising information for capturing phenological patterns. Based on those results, we suggest that individuals from the same species and functional group might be identified using digital images, and introduce a new tool to help phenology experts in the identification of new individuals from the same species in the image and their location on the ground. © 2013 Elsevier B.V. All rights reserved.

Associação entre aditivos químicos e bacterianos na ensilagem de cana-de-açúcar

Objetivou-se avaliar a ensilagem de cana-de-açúcar tratada com três aditivos químicos (uréia 1,5%, benzoato de sódio 0,1% e hidróxido de sódio 1%) mais o grupo controle e duas inoculações (Propionibacterium acidipropionici + Lactobacillus plantarum e Lactobacillus buchneri), em um esquema fatorial 4 x 3 com três repetições para cada tratamento. Avaliou-se o valor nutritivo da forragem antes da ensilagem, após a abertura dos silos e após a exposição aeróbia. As associações de P. acidipropionici ou L. buchneri com NaOH, em comparação ao grupo controle, possibilitaram melhor preservação dos teores de MS (32,2 e 33,5 vs 27,4%, respectivamente), FDN ( 53,4; 55,7 vs 75,3%), FDA (39,5; 44,3 vs 48,7%), lignina (6,6; 7,1 vs 8,1%) e CNF (33,8; 31,7 vs 14,9%) e, conseqüentemente, propiciaram os maiores valores de DIVMS (60,3; 63,2 vs 35,1%). Esses valores podem ser atribuídos ao controle das leveduras pelos efeitos da associação dos aditivos. A ensilagem de cana-de-açúcar requer de forma contundente a inclusão de aditivo.

EFFICIENT FAULT LOCATION IN UNDERGROUND DISTRIBUTION SYSTEMS THROUGH OPTIMUM-PATH FOREST

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

Efficient supervised optimum-path forest classification for large datasets

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

HOW FAR do WE GET USING MACHINE LEARNING BLACK-BOXES?

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

An Intelligent System for Petroleum Well Drilling Cutting Analysis

Cutting analysis is a important and crucial task task to detect and prevent problems during the petroleum well drilling process. Several studies have been developed for drilling inspection, but none of them takes care about analysing the generated cutting at the vibrating shale shakers. Here we proposed a system to analyse the cutting's concentration at the vibrating shale shakers, which can indicate problems during the petroleum well drilling process, such that the collapse of the well borehole walls. Cutting's images are acquired and sent to the data analysis module, which has as the main goal to extract features and to classify frames according to one of three previously classes of cutting's volume. A collection of supervised classifiers were applied in order to allow comparisons about their accuracy and efficiency. We used the Optimum-Path Forest (OPF), Artificial Neural Network using Multi layer Perceptrons (ANN-MLP), Support Vector Machines (SVM) and a Bayesian Classifier (BC) for this task. The first one outperformed all the remaining classifiers. Recall that we are also the first to introduce the OPF classifier in this field of knowledge. Very good results show the robustness of the proposed system, which can be also integrated with other commonly system (Mud-Logging) in order to improve the last one's efficiency.

HIGH STRENGTH CONCRETE: THE REINFORCED CONCRETE OF THE CENTURY 21(ST)

This work is about the 21st century reinforced concrete analysis under the point of view of its constituent materials. First of all it is described the theoretical approach of the bending elements calculated based on the Norms BAEL 91 standarts. After that, numerical load-displacement are presented from reinforced concrete beams and plates validated by experimental data. The numerical modellings has been carried on in the program CASTEM 2000. In this program a elastoplastic model of Drucker-Prager defines the rupture surface of the concrete in non associative plasticity. The crack is smeared on the Gauss points of the finite elements with formation criterion starting from the definition of the rupture surface in the branch traction-traction of the Rankine model. The reinforcements were modeled in a discrete approach with perfect bond. Finally, a comparative analysis is made between the numerical results and calculated criteria showing the future of high performance reinforced concrete in this beginning of 21st century.

Repeated unrewarded scent exposure influences the food choice of stingless bee foragers, Melipona scutellaris

Conselho Nacional de Desenvolvimento Científico e Tecnológico (CNPq)

Multiscale Fractal Descriptors and Polynomial Classifier for Partial Pixels Identification in Regions of Interest of Mammographic Images

Computer systems are used to support breast cancer diagnosis, with decisions taken from measurements carried out in regions of interest (ROIs). We show that support decisions obtained from square or rectangular ROIs can to include background regions with different behavior of healthy or diseased tissues. In this study, the background regions were identified as Partial Pixels (PP), obtained with a multilevel method of segmentation based on maximum entropy. The behaviors of healthy, diseased and partial tissues were quantified by fractal dimension and multiscale lacunarity, calculated through signatures of textures. The separability of groups was achieved using a polynomial classifier. The polynomials have powerful approximation properties as classifiers to treat characteristics linearly separable or not. This proposed method allowed quantifying the ROIs investigated and demonstrated that different behaviors are obtained, with distinctions of 90% for images obtained in the Cranio-caudal (CC) and Mediolateral Oblique (MLO) views.

On the efficacy of texture analysis for crowd monitoring

The goal of this work is to assess the efficacy of texture measures for estimating levels of crowd densities ill images. This estimation is crucial for the problem of crowd monitoring. and control. The assessment is carried out oil a set of nearly 300 real images captured from Liverpool Street Train Station. London, UK using texture measures extracted from the images through the following four different methods: gray level dependence matrices, straight lille segments. Fourier analysis. and fractal dimensions. The estimations of dowel densities are given in terms of the classification of the input images ill five classes of densities (very low, low. moderate. high and very high). Three types of classifiers are used: neural (implemented according to the Kohonen model). Bayesian. and an approach based on fitting functions. The results obtained by these three classifiers. using the four texture measures. allowed the conclusion that, for the problem of crowd density estimation. texture analysis is very effective.

Petroleum well drilling monitoring through cutting image analysis and artificial intelligence techniques

Petroleum well drilling monitoring has become an important tool for detecting and preventing problems during the well drilling process. In this paper, we propose to assist the drilling process by analyzing the cutting images at the vibrating shake shaker, in which different concentrations of cuttings can indicate possible problems, such as the collapse of the well borehole walls. In such a way, we present here an innovative computer vision system composed by a real time cutting volume estimator addressed by support vector regression. As far we know, we are the first to propose the petroleum well drilling monitoring by cutting image analysis. We also applied a collection of supervised classifiers for cutting volume classification. (C) 2010 Elsevier Ltd. All rights reserved.

Estimativa de ocorrência de precipitação em áreas agrícolas utilizando floresta de caminhos ótimos

As condições meteorológicas são determinantes para a produção agrícola; a precipitação, em particular, pode ser citada como a mais influente por sua relação direta com o balanço hídrico. Neste sentido, modelos agrometeorológicos, os quais se baseiam nas respostas das culturas às condições meteorológicas, vêm sendo cada vez mais utilizados para a estimativa de rendimentos agrícolas. Devido às dificuldades de obtenção de dados para abastecer tais modelos, métodos de estimativa de precipitação utilizando imagens dos canais espectrais dos satélites meteorológicos têm sido empregados para esta finalidade. O presente trabalho tem por objetivo utilizar o classificador de padrões floresta de caminhos ótimos para correlacionar informações disponíveis no canal espectral infravermelho do satélite meteorológico GOES-12 com a refletividade obtida pelo radar do IPMET/UNESP localizado no município de Bauru, visando o desenvolvimento de um modelo para a detecção de ocorrência de precipitação. Nos experimentos foram comparados quatro algoritmos de classificação: redes neurais artificiais (ANN), k-vizinhos mais próximos (k-NN), máquinas de vetores de suporte (SVM) e floresta de caminhos ótimos (OPF). Este último obteve melhor resultado, tanto em eficiência quanto em precisão.

Ruminal parameters analyzed in remaining digestion residue of roughages in the in vitro/gas system

O desempenho animal é a medida mais direta de se avaliar a qualidade dos alimentos. Entretanto, dados de desempenho são insuficientes para se detectar as possíveis interações que possam ocorrer no ambiente ruminal. O objetivo do presente trabalho foi avaliar os possíveis efeitos associativos nas concentrações de ácidos graxos voláteis (AGVs), nitrogênio amoniacal (N-NH3) e pH da fração líquida remanescente da digestão da matéria seca (MS) de volumosos exclusivos (cana-de-açúcar= CN; capim-elefante com 60 dias= CP60 e 180 dias= CP180 de crescimento; e silagem de milho= SIL) e suas combinações (cana-de-açúcar+silagem de milho= CNSIL; cana-de-açúcar+capim-elefante-60d= CNCP60; cana-de-açúcar+capim-elefante-180d= CNCP180; silagem de milho+capim-elefante-60d= SILCP60; silagem de milho+capim-elefante-180d= SILCP180) na proporção de 50% na MS, que levam a resultados de desempenhos positivos ou negativos de bovinos. As concentrações de AGVs, N-NH3 e pH dos tratamentos foram: CN= 56,9 mmol L-1, 50,1 mg dL-1, 5,7; CNSIL= 61,4 mmol L-1, 50,7 mg dL-1, 5,8; CNCP60= 54,7 mmol L-1, 47,6 mg dL-1, 5,8; CNCP180= 45,4 mmol L-1, 49,4 mg dL-1, 6,0; SIL= 57,2 mmol L-1, 54,0 mg dL-1, 5,8; SILCP60= 57,1 mmol L-1, 53,1 mg dL-1, 5,9; SILCP180= 55,9 mmol L-1, 52,3 mg dL-1, 6,0; CP60= 58,1 mmol L-1, 49,4 mg dL-1, 5,9; CP180= 44,0 mmol L-1, 46,4 mg dL-1, 6,1. Os carboidratos não estruturais e amido, aliados à fibra e proteína, contribuíram para que ocorresse o efeito associativo positivo na mistura 50:50 cana/silagem. Isso pode ter propiciado os melhores resultados de desempenho em bovinos devido ao elevado padrão fermentativo.