Embedded pattern recognition systems for liquids classification: A comparison study

Author

Gil-Sánchez, Luis ; Garcia-Breijo, Eduardo ; Garrigues, José ; Laguarda, Nicolás ; Masot, Rafael ; Ibáñez, Javier ; Atkinson, John ; Glanc, Monika

Author_Institution

Centro de Reconocimiento Mol. y Desarrollo Tecnol., Univ. Politec. de Valencia, Valencia, Spain

fYear

2011

fDate

28-31 Oct. 2011

Firstpage

1720

Lastpage

1723

Abstract

A comparative study among a Fuzzy ARTMAP, a Multi-Layer Feed-Forward network (MLFF) and a Linear Discriminant Analysis (LDA) has been done in a liquids classification system in order to obtain the best implementation on a microcontroller based system. These algorithms are initially trained with a multi-electrode array on a Personal Computer (PC) using several samples of waters to obtain the optimum architecture of the networks. Once it is trained, the computed data are programmed into the microcontroller, which then gives the water classification directly for new unknown water samples. The multi-electrode has been developed using thick-film technology.

Keywords

chemical engineering computing; electrodes; embedded systems; feedforward neural nets; fuzzy neural nets; microcomputers; microcontrollers; pattern classification; thick films; water; water quality; embedded pattern recognition system; fuzzy ARTMAP; linear discriminant analysis; liquid classification system; microcontroller based system; multielectrode array; multilayer feedforward network; personal computer; thick-film technology; water classification system; Algorithm design and analysis; Arrays; Electrodes; Microcontrollers; Pattern recognition; Random access memory; Training;

fLanguage

English

Publisher

ieee

Conference_Titel

Sensors, 2011 IEEE

Conference_Location

Limerick

ISSN

1930-0395

Print_ISBN

978-1-4244-9290-9

Type

conf

DOI

10.1109/ICSENS.2011.6126921

Filename

6126921