Rolf Fredi Molz

Propose of a Hardware Implementation for Fingerprint Systems

Fingerprint is graphical flow-like ridges presents on human fingers. Each fingerprint is unique, offering a clear and unambiguous method to identify an individual. The uniqueness of each fingerprint is determined by fine details embedded in its overall structure, named minutiae. Fingerprint classification system is CPU time intensive, usually implemented in software. This paper presents an alternative way to identify the minutiae from fingerprints, aiming real-time processing. In the first part is implemented the alternative algorithm in software (Delphi) and after this is presented an architecture to be implemented using a configurable devices (FPGA). The performance of this algorithm, in hardware and software, are analyzed, presenting the spent time within each system block.

A new proposal for implementation of competitive neural networks in analog hardware

This paper proposes the basic structure of a competitive neural network (CNN), based on analog IC techniques, for hardware implementation. This leads to a more compact project and allows real time processing. It is shown that the fundamental equations that describe the behavior of competitive neural networks possess a relationship with some basic electronic components. This fact allows the direct implementation of CNN with these electronic components. Initially the behavior of the fundamental equations of this type of neural networks is studied by means of software simulations. This behavior is then compared with the one obtained through electric simulations of the equivalent circuits originated from these fundamental equations. Simulations show that the most important features of the CNN are obtained with the presented implementation. Finally, a typical application is presented in the area of pattern clustering using synaptic weights to demonstrate an implementation using the techniques described.

Análise exploratória aplicada no estudo de medicamentos contendo piroxicam

The identification of different pharmaceutical formulations with piroxicam was studied, using spectra of diffuse reflectance infrared Fourier transform spectroscopy (DRIFTS), in association with hierarchical cluster analysis (HCA) technique. The spectra of samples of the 5 different compounding pharmacies, containing 10 or 20 mg of piroxicam and its respective inactive ingredients, had been collected in Nicolet Magna 550 spectrophotometer. For the multivariate analysis, the spectral information had been processed in software package Pirouette® 2.7 of the Infometrix. The dendograms had been constructed with the autoscaled data, and multiplicative scatter correction (MSC), using three types of linkage methods: single, flexible and incremental. By applying the hierarchical cluster analysis the formation of two distinct groups: one with the active principles and another group with the inactive ingredients was proved. These results demonstrate that the DRIFTS associated with chemometrics tools, constitute a quality control alternative of drugs production processes.

Determination of amoxicillin content in powdered pharmaceutical formulations using DRIFTS and PLS

The amount of amoxicillin in pharmaceutical formulations was determined using spectra of diffuse reflectance infrared Fourier transform spectroscopy (DRIFTS), in association with partial least squares (PLS) regression. The spectra of 24 different samples (17 for the calibration set and 7 for the validation set), which had been registered using a Nicolet Magna 550 spectrophotometer, contained 76.7-94.3% of amoxicillin in starch. The PLS models were constructed with auto-scaled or variance-scaled data, and multiplicative scatter correction (MSC). The best model presented R2=0.9936, RMSEC=0.441 and RMSEV=0.790. The analytical method proposed is non-destructive and the cost and time of analysis are very much reduced, allowing for the fast and direct determination of amoxicillin content.

Uso de um Ambiente Codesign para a Implementação de Redes Neurais

This paper presents a hardware implementation of neural networks into reprogrammable devices, FPGAs. It was used for this implementation the LIRMM board, which allows to develop both hardware and software designs, using C and VHDL languages, respectively. A feedforward neural network was implemented, using backpropagation learning. Emphasis is given for the net calculus, since intensive multiplications and additions are needed to compute this function. At the neuron level, the size of arithmetic operators was reduced by correctly coding the synaptics weights and input patterns, as well as replacing multipliers by adders and look-up-tables. At the network level, two techniques were employed to neuron synchronization. The synchronous solution resulted in a correct neural network implemented in hardware. After this implementation, the simulation of the neural network is presented. This work is organized as follows: section 1 introduces the use of neural network; section 2 presents the hardware-software environment used; section 3 details the neural network implementations (digital, analog, hybrid), our network and the simulation; and finally, our conclusions and future work.

Otimização de métodos de controle de qualidade de fármacos usando algoritmo genético e busca tabu

Abstract Two new methodologies of analysis for the antihipertensive drugs quantification are provided in this work, using meta-heuristic algorithms optimization with infrared spectroscopy by diffuse reflectance Fourier transform (DRIFTS) and partial-least squares regression method (PLS). A diffuse reflectance joined to a spectrophotometry in the mid-infrared region was used. The calibration was through PLS in order to use it as a pre-processing. The data were used in multiplicative signal correction (MSC). After the best wavelength selection, which results in models with better coefficient regression (R 2 ), these were improved using routines based on genetic algorithm heuristic and tabu search methods. For the evaluation of the optimization a standard error of validation (SEV) was used. The new technique using DRIFTS/PLS/AG or DRIFTS/PLS/TS shows excellent choices for the quality control of the productive process of factories and drugstore. Keywords: genetic algorithm; tabu search; chemometrics; meta-heuristics.

Design of a Classification System for Rectangular Shapes Using a Co-Design Environment

Pattern localization and classification are CPU time intensive, being normally implemented in software. Custom implementations in hardware allow real-time processing. In practice, in ASIC or FPGA implementations, the digitization process introduces errors that should be taken into account. This paper presents initially the state-of-the-art in this field, analyzing the performance and implementation of each work. After we propose a system for rectangular shapes localization and classification using reconfigurable devices (FPGA) and a signal processor (DSP) available in a flexible codesign platform. The system will be described using C and VHDL languages, for the software and hardware parts respectively. Finally, the classification block of the system, implemented as an artificial neural network (ANN) in a rapid prototyping platform, is described.

Desenvolvimento e Avaliação de um Protótipo de Colorímetro por Imagens Digitais

This work aimed to developed a prototype system for colorimetric determinations by digital images, as well evaluate their functionality to determine phosphorus concentration in natural waters. This study was motivated by the necessity of portable tools, alternatively to spectrometer, to optimize the analysis time, transportation costs and storage of samples. The results of the proposed device were compared with a spectrophotometric reference method, and the proposed method presented satisfactory results for phosphorus determination.

Simulador para Introdução da Programação para Crianças e Análise da Aprendizagem com apoio da Neurociência

A programacao requer um alto nivel de atencao, o que para os adultos ja nao e tarefa facil. Para criancas, inseridas em um mundo virtual imagetico, intuitivo e animado, a programacao necessita ser introduzida seguindo estas caracteristicas. Neste sentido, foi desenvolvido um simulador on-line interativo que permite as criancas o controle de motores e leds por programacao. Inicialmente, as criancas utilizaram o kit fisico, ja que a manipulacao de material concreto facilita o processo de abstracao necessario a programacao. Em um segundo momento, foi introduzido o simulador, e, em ambos os casos, a interacao foi satisfatoria. O simulador traz como vantagem o seu uso por meio de uma interface web, possibilitando as criancas continuarem suas atividades extraclasses. Para realizar a analise da aprendizagem, duas etapas foram constituidas: a primeira foi referente a validacao do simulador, bem como observacao do emprego da memoria para coordenar a programacao; na segunda, serao utilizados o Metodo Clinico de Piaget em comparacao com a analise da evocacao da memoria explicita (coletada na primeira etapa), relacionada ao estudo da neurociencia. Este artigo apresenta a primeira etapa do projeto.