Desenvolvimento de sensores descartáveis de ouro nanoestruturado modificados com complexo metálico visando detecção in situ de corantes semipermanentes de cabelo

Pós-graduação em Química - IQ

Aplicação de sistemas neuro-fuzzy e espectrometria no infravermelho próximo para a identificação em tempo real do teor de nitrogênio foliar em cana-de-açúcar

Pós-graduação em Engenharia Elétrica - FEB

Structural, dielectric, ferroelectric and optical properties of PBCT, PBST and PCST complex thin films on LaNiO3 metallic conductive oxide layer coated Si substrates by the CSD technique

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

Pruning methods to MLP neural networks considering proportional apparent error rate for classification problems with unbalanced data

This article deals with classification problems involving unequal probabilities in each class and discusses metrics to systems that use multilayer perceptrons neural networks (MLP) for the task of classifying new patterns. In addition we propose three new pruning methods that were compared to other seven existing methods in the literature for MLP networks. All pruning algorithms presented in this paper have been modified by the authors to do pruning of neurons, in order to produce fully connected MLP networks but being small in its intermediary layer. Experiments were carried out involving the E. coli unbalanced classification problem and ten pruning methods. The proposed methods had obtained good results, actually, better results than another pruning methods previously defined at the MLP neural network area. (C) 2014 Elsevier Ltd. All rights reserved.

Social-spider optimization-based artificial neural networks training and its applications for Parkinson's disease identification

Evolutionary algorithms have been widely used for Artificial Neural Networks (ANN) training, being the idea to update the neurons' weights using social dynamics of living organisms in order to decrease the classification error. In this paper, we have introduced Social-Spider Optimization to improve the training phase of ANN with Multilayer perceptrons, and we validated the proposed approach in the context of Parkinson's Disease recognition. The experimental section has been carried out against with five other well-known meta-heuristics techniques, and it has shown SSO can be a suitable approach for ANN-MLP training step.

Optimum-Path Forest Applied for Breast Masses Classification

In Computer-Aided Diagnosis-based schemes in mammography analysis each module is interconnected, which directly affects the system operation as a whole. The identification of mammograms with and without masses is highly needed to reduce the false positive rates regarding the automatic selection of regions of interest for further image segmentation. This study aims to evaluate the performance of three techniques in classifying regions of interest as containing masses or without masses (without clinical findings), as well as the main contribution of this work is to introduce the Optimum-Path Forest (OPF) classifier in this context, which has never been done so far. Thus, we have compared OPF against with two sorts of neural networks in a private dataset composed by 120 images: Radial Basis Function and Multilayer Perceptron (MLP). Texture features have been used for such purpose, and the experiments have demonstrated that MLP networks have been slightly better than OPF, but the former is much faster, which can be a suitable tool for real-time recognition systems.

Aplicabilidade das equações isotermas de Freundlich e Langmuir na interação biossortiva de corantes têxteis com Saccharomyces cerevisiae liofilizada

This paper presents a study of the applicability of adsorption isotherms, known as Langmuir and Freundlich isotherm, between the biosorptive interaction of yeast lyophilized Saccharomyces cerevisiae and textile dyes. To that end, we prepared stock solutions of the textile dyes Direct Red 23 and Direct Red 75 in the concentration of 1.000μg/mL and a yeast suspension at 2,5%. We did experiments for two cases, firstly for the case that we have a fix concentration of yeast at 0,500mg/mL and an variable concentration of dye range from40, 50, 60, 80 and 100μg/mL, then for the case that we fixed the concentration of dye at 100μg/mL and the yeast concentration was variable range from 0,250, 0,500, 0,750, 1,000, 1,250mg/mL. For the dye Direct Red 23 we did analysis in the pH 2,5, 4,5 and 6,5; for the Direct Red 75, we just did for the pH 2,5. We leave the dye solution in contact with the yeast for 2 hours at a constant temperature of 30°C and then centrifuged and analyzed the sample in a spectrophotometer and finally made and analysis of parameters for the removal and study of the isotherms. After the biosorption, was observed that for the Direct Red 23 in the pH 2,5 was needed 1,407mg/mL of yeast for total removal, while for the pH 4,5 was needed 8,806mg/mL and in pH 6,5 was 9,286mg/mL; for the Direct Red 75 in pH 2,5 was needed 1,337mg/mL. This difference can be explain by the adsorption isotherms, was observed that in the case when the yeast was fix when we had in a acid pH the behavior of the system was compatible with the Langmuir isotherm, and thus, an monolayer pattern. And that when we decrease the acidity of the medium the system became more compatible with a Freundlich isotherm, and thus, a multilayer pattern; for the case that the yeast was variable this is not much evident, however for the pH 2,5 she became compatible with a Langmuir isotherm... (Complete abstract click electronic access below)

Contagem automatizada de veículos utilizando segmentação estatística

In this work was developed a program capable of performing automatic counting of vehicles on roads. The problem of counting vehicles is using expensive techniques for its realization, techniques which often involve manual counting or degradation of the pavement. The main motivation for this work was the importance that the vehicle counting represents to the Traffic Engineer, being essential to analyze the performance of the roads, allowing to measure the need for installation of traffic lights, roundabouts, access ways, among other means capable of ensuring a continuous flow and safe for vehicles. The main objective of this work was to apply a statistical segmentation technique recently developed, based on a nonparametric linear regression model, to solve the segmentation problem of the program counter. The development program was based on the creation of three major modules, one for the segmentation, another for the tracking and another for the recognition. For the development of the segmentation module, it was applied a statistical technique combined with the segmentation by background difference, in order to optimize the process. The tracking module was developed based on the use of Kalman filters and application of simple concepts of analytical geometry. To develop the recognition module, it was used Fourier descriptors and a neural network multilayer perceptron, trained by backpropagation. Besides the development of the modules, it was also developed a control logic capable of performing the interconnection among the modules, mainly based on a data structure called state. The analysis of the results was applied to the program counter and its component modules, and the individual analysis served as a means to establish the par ameter values of techniques used. The find result was positive, since the statistical segmentation technique proved to be very useful and the developed program was able to count the vehicles belonging to the three goal..

Segurança em redes sem fio: estudo sobre o desenvolvimento de conjuntos de dados para comparação de IDS

Pós-graduação em Engenharia Elétrica - FEIS

Inspeção de materiais compósitos utilizando ondas acústicas guiadas

Pós-graduação em Engenharia Elétrica - FEIS

Neural tool condition estimation in the grinding of advanced ceramics

Ceramic parts are increasingly replacing metal parts due to their excellent physical, chemical and mechanical properties, however they also make them difficult to manufacture by traditional machining methods. The developments carried out in this work are used to estimate tool wear during the grinding of advanced ceramics. The learning process was fed with data collected from a surface grinding machine with tangential diamond wheel and alumina ceramic test specimens, in three cutting configurations: with depths of cut of 120 mu m, 70 mu m and 20 mu m. The grinding wheel speed was 35m/s and the table speed 2.3m/s. Four neural models were evaluated, namely: Multilayer Perceptron, Radial Basis Function, Generalized Regression Neural Networks and the Adaptive Neuro-Fuzzy Inference System. The models'performance evaluation routines were executed automatically, testing all the possible combinations of inputs, number of neurons, number of layers, and spreading. The computational results reveal that the neural models were highly successful in estimating tool wear, since the errors were lower than 4%.

Tool condition monitoring of aluminum oxide grinding wheel in dressing operation using acoustic emission and neural networks

The grinding operation gives workpieces their final finish, minimizing surface roughness through the interaction between the abrasive grains of a tool (grinding wheel) and the workpiece. However, excessive grinding wheel wear due to friction renders the tool unsuitable for further use, thus requiring the dressing operation to remove and/or sharpen the cutting edges of the worn grains to render them reusable. The purpose of this study was to monitor the dressing operation using the acoustic emission (AE) signal and statistics derived from this signal, classifying the grinding wheel as sharp or dull by means of artificial neural networks. An aluminum oxide wheel installed on a surface grinding machine, a signal acquisition system, and a single-point dresser were used in the experiments. Tests were performed varying overlap ratios and dressing depths. The root mean square values and two additional statistics were calculated based on the raw AE data. A multilayer perceptron neural network was used with the Levenberg-Marquardt learning algorithm, whose inputs were the aforementioned statistics. The results indicate that this method was successful in classifying the conditions of the grinding wheel in the dressing process, identifying the tool as "sharp''(with cutting capacity) or "dull''(with loss of cutting capacity), thus reducing the time and cost of the operation and minimizing excessive removal of abrasive material from the grinding wheel.

Characterization and application of nanostructured films containing Au and TiO2 nanoparticles supported in bacterial cellulose

Conselho Nacional de Desenvolvimento Científico e Tecnológico (CNPq)

Comparative study between powers of sigmoid functions, MLP-backpropagation and polynomials in function approximation problems

Function approximation is a very important task in environments where the computation has to be based on extracting information from data samples in real world processes. So, the development of new mathematical model is a very important activity to guarantee the evolution of the function approximation area. In this sense, we will present the Polynomials Powers of Sigmoid (PPS) as a linear neural network. In this paper, we will introduce one series of practical results for the Polynomials Powers of Sigmoid, where we will show some advantages of the use of the powers of sigmiod functions in relationship the traditional MLP-Backpropagation and Polynomials in functions approximation problems.

Utilização de redes neurais artificiais aplicadas na discriminação de padrões de doenças florestais

Pós-graduação em Ciência Florestal - FCA