Oriented texture classification based on self-organizing neural network and Hough transform

This paper presents a technique for oriented texture classification which is based on the Hough transform and Kohonen's neural network model. In this technique, oriented texture features are extracted from the Hough space by means of two distinct strategies. While the first operates on a non-uniformly sampled Hough space, the second concentrates on the peaks produced in the Hough space. The described technique gives good results for the classification of oriented textures, a common phenomenon in nature underlying an important class of images. Experimental results are presented to demonstrate the performance of the new technique in comparison, with an implemented technique based on Gabor filters.

Activation function study for wavelet network

The main purpose of this paper is to investigate theoretically and experimentally the use of family of Polynomial Powers of the Sigmoid (PPS) Function Networks applied in speech signal representation and function approximation. This paper carries out practical investigations in terms of approximation fitness (LSE), time consuming (CPU Time), computational complexity (FLOP) and representation power (Number of Activation Function) for different PPS activation functions. We expected that different activation functions can provide performance variations and further investigations will guide us towards a class of mappings associating the best activation function to solve a class of problems under certain criteria.

Constraints implementation for IQML and mode direction-of-arrival estimators

The iterative quadratic maximum likelihood IQML and the method of direction estimation MODE are well known high resolution direction-of-arrival DOA estimation methods. Their solutions lead to an optimization problem with constraints. The usual linear constraint presents a poor performance for certain DOA values. This work proposes a new linear constraint applicable to both DOA methods and compare their performance with two others: unit norm and usual linear constraint. It is shown that the proposed alternative performs better than others constraints. The resulting computational complexity is also investigated.

Electromyographic validation of the trapezius and serratus anterior muscles in military press exercises with middle grip

Electromyographic activity of the trapezius muscle and serratus anterior muscle was analysed in 24 male volunteers using a 2-channel TECA TE 4 electromyograp, during the execution of four different modalities of military press exercises with middle grip. The trapezius acted preferentially in the modalities standing press behind neck; and sitting forward and press behind neck, while SI did not show any significative difference among the modalities. The high levels of action potentials with which TS and SI acted justify the inclusion of these exercises in physical programmes.

Evolvable hardware applied to voice recognition

This paper presents some results of the application on Evolvable Hardware (EHW) in the area of voice recognition. Evolvable Hardware is able to change inner connections, using genetic learning techniques, adapting its own functionality to external condition changing. This technique became feasible by the improvement of the Programmable Logic Devices. Nowadays, it is possible to have, in a single device, the ability to change, on-line and in real-time, part of its own circuit. This work proposes a reconfigurable architecture of a system that is able to receive voice commands to execute special tasks as, to help handicapped persons in their daily home routines. The idea is to collect several voice samples, process them through algorithms based on Mel - Ceptrais theory to obtain their numerical coefficients for each sample, which, compose the universe of search used by genetic algorithm. The voice patterns considered, are limited to seven sustained Portuguese vowel phonemes (a, eh, e, i, oh, o, u).

Effects of the thyristor controlled serie capacitor on the electric power system

This work deals with the effects of the series compensation on the electric power system for small-signal stability studies. Therefore, the system is modeled admitting the existence of the compensation and then, the equations are linearized and a linear model is obtained for a single machine-infinite bus power system with a compensator installed. The resulting model with nine defined constants is very similar to the Heffron & Phillips linear model widely used on the existent literature. Finally, simulations are executed for an example system, to analyze the behavior of these constants when loading the system. © 2004 IEEE.

Virtual instrumentation applied to the implementation of IEEE STD 1459-2000 power definitions

This paper discusses the utilization of Virtual Instrumentation to the implementation and evaluation of different power definitions, so that classical formulations and new definitions can be compared without the necessity of acquiring different power meters or analyzers. Accordingly, the definitions of IEEE Standard 1459-2000 for the measurement of power quantities under distorted and unbalanced situations, have been digitally implemented. Thus, several power and power factor components related to the decomposition of the measured voltage and current signals have been obtained. The proposed PC-based Virtual Instrument uses a high performance acquisition board and isolated sensors and transducers. All digital algorithms and routines have been implemented by means of a graphical development system. Regarding to the implementation of STD 1459, this paper also proposes several different algorithms to the required decompositions of voltage, current and power components. © 2005 IEEE.

In-process grinding monitoring through acoustic emission

This work aims to investigate the efficiency of digital signal processing tools of acoustic emission signals in order to detect thermal damages in grinding processes. To accomplish such a goal, an experimental work was carried out for 15 runs in a surface grinding machine operating with an aluminum oxide grinding wheel and ABNT 1045 Steel as work material. The acoustic emission signals were acquired from a fixed sensor placed on the workpiece holder. A high sampling rate data acquisition system working at 2.5 MHz was used to collect the raw acoustic emission instead of the root mean square value usually employed. Many statistical analyses have shown to be effective to detect burn, such as the root mean square (RMS), correlation of the AE, constant false alarm rate (CFAR), ratio of power (ROP) and mean-value deviance (MVD). However, the CFAR, ROP, Kurtosis and correlation of the AE have been presented more sensitive than the RMS. Copyright © 2006 by ABCM.

Oropharyngeal dysphagia identification using wavelets and optimum path forest

The swallowing disturbers are defined as oropharyngeal dysphagia when present specifies signals and symptoms that are characterized for alterations in any phases of swallowing. Early diagnosis is crucial for the prognosis of patients with dysphagia and the potential to diagnose dysphagia in a noninvasive manner by assessing the sounds of swallowing is a highly attractive option for the dysphagia clinician. This study proposes a new framework for oropharyngeal dysphagia identification, having two main contributions: a new set of features extract from swallowing signal by discrete wavelet transform and the dysphagia classification by a novel pattern classifier called OPF. We also employed the well known SVM algorithm in the dysphagia identification task, for comparison purposes. We performed the experiments in two sub-signals: the first was the moment of the maximal peak (MP) of the signal and the second is the swallowing apnea period (SAP). The OPF final accuracy obtained were 85.2% and 80.2% for the analyzed signals MP and SAP, respectively, outperforming the SVM results. ©2008 IEEE.

Mandible protrusion and decrease of TMJ sounds: An electrovibratographic examination

This study quantified by, electrovibratography, the amount of mandible protrusion required to decrease significantly temporomandibular joint (TMJ) vibratory energy as an aid in the diagnosis of the recapture of anteriorly displaced disk. Eighteen patients diagnosed as having anterior disk displacement with reduction and TMJ clicking were submitted to electrovibratographic examination at the first appointment and treated with a stabilizing appliance and anterior positioning appliance with 1 to 5 mm protrusion. Vibratory energy was checked in each of these positions. Baseline data were used as control. At the first appointment, the patients had vibrations with more elevated intensities at the middle and late phases of the mouth opening cycle. At only one clinical step, mandible protrusion was obtained with the anterior repositioning appliance, ranging from 1 to 5 mm protusion. At each new position, a new electrovibratographic exam was made. After the 5-mm mandibular projection, only 2 patients presented vibration, with means between 0.6 and 2.8 Hz. Data were analyzed statistically by ANOVA and Tukey's test (α=0.05). The outcomes of this study indicate that 3 mm is the minimum amount of mandible protrusion to significantly decrease the TMJ vibratory energy and to recapture the displaced articular disk.

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.

Methodology proposal for acquisition of biotelemetric data in a real time

This work presents a methodological proposal for acquisition of biometric data through telemetry basing its development on a research-action and a case study. Nowadays, the qualified professionals of physical evaluation have to use specific devices to obtain biometric signals and data. These devices in the most of the time are high cost and difficult to use and handling. Therefore, the methodological proposal was elaborate in order to develop, conceptually, a bio telemetric device which could acquire the desirable biometric signals: oxymetry, biometrics, corporal temperature and pedometry which are essential for the area of physical evaluation. It was researched the existent biometrics sensors, the possible ways for the remote transmission of signals and the computer systems available so that the acquisition of data could be possible. This methodological proposal of remote acquisition of biometrical signals is structured in four modules: Acquisitor of biometrics data; Converser and transmitter of biometric signals; Receiver and Processor of biometrics signals and Generator of Interpretative Graphs. The modules aim the obtention of interpretative graphics of human biometric signals. In order to validate this proposal a functional prototype was developed and it is presented in the development of this work.

Conversor CA-CA de elevado fator de potência para aplicações com geradores de Ímã permanente

This paper proposes and describes a high power factor AC-AC converter for naval applications using Permanent Magnet Generator (PMG). The three-phase output voltages of the PMG vary from 260 Vrms (220 Hz) to 380 Vrms (360 Hz), depending on load conditions. The proposed converter consists of a Y-/ΔY power transformer, which provides electrical isolation between the PMG and remaining stages, and a twelve-pulse uncontrolled rectifier stage directly connected to a single-phase inverter stage, without the use of an intermediary DC-DC topology. This proposal results in more simplicity for the overall circuitry, assuring robustness, reliability and reduced costs. Furthermore, the multipulse rectifier stage is capable to provide high power factor and low total harmonic distortion for the input currents of the converter. The single-phase inverter stage was designed to operate with wide range of DC bus voltage, maintaining 120 Vrms, 60 Hz output. The control philosophy, implemented in a digital signal processor (DSP) which also contains protection routines, alows series connections between two identical converters, achieving 240 Vrms, 60 Hz total output voltage. Measured total harmonic distortion for the AC output voltage is lower than 2% and the input power factor is 0.93 at 3.6kW nominal load. © 2010 IEEE.

ENCORE: Enhanced code Galileo receiver for land management applications in Brazil

Taking benefit of the new Galileo ranging signals, the ENCORE (Enhanced Code Galileo Receiver) project aims to develop a low-cost Land Management Application to cover needs of the Brazilian market in terms of geo-referencing and rural/urban cadastre, using a low-cost Enhanced Galileo Code Receiver as baseline. Land management applications require precision and accuracy levels from a few to several decimetres that are under-met with current pseudorange-based receiver and over-met with phase observations. This situation leads either to a waste of resources, or to lack of accuracy. In this project, it is proposed to fill this gap using the new possibilities of the Galileo ranging signals, in particular E5 AltBOC and E1 CBOC. This approach reduces the cost of the end-user solution, helping the rapid penetration of Galileo technology outside Europe. ©2010 IEEE.

Preprocessing data for short-term load forecasting with a general regression neural network and a moving average filter

This paper proposes a filter based on a general regression neural network and a moving average filter, for preprocessing half-hourly load data for short-term multinodal load forecasting, discussed in another paper. Tests made with half-hourly load data from nine New Zealand electrical substations demonstrate that this filter is able to handle noise, missing data and abnormal data. © 2011 IEEE.