Neural network and pattern recognition techniques for characterizing ultrasonic transducers

Author

Obaidat, M.S. ; Abu-Saymeh, D.S.

Author_Institution

Dept. of Electr. & Comput. Eng., Missouri Univ., Columbia, MO, USA

fYear

1992

fDate

1-3 April 1992

Firstpage

729

Lastpage

735

Abstract

A system in which transducers are characterized and classified using both pattern recognition and neural network algorithms is presented. Various techniques are investigated, compared, and analyzed. The hardware of the system used in collecting and measuring the characteristics of the transducers, and the parameters used in classifying the transducers into classes are described. The neural network and pattern recognition algorithms used to classify the transducers are introduced. The results of the classification are presented and the various algorithms are analyzed and compared.<>

Keywords

neural nets; pattern recognition; ultrasonic transducers; characterizing ultrasonic transducers; neural network techniques; pattern recognition; Application software; Artificial neural networks; Computer errors; Computer networks; Hardware; Multi-layer neural network; Neural networks; Pattern recognition; Ultrasonic transducers; Ultrasonic variables measurement;

fLanguage

English

Publisher

ieee

Conference_Titel

Computers and Communications, 1992. Conference Proceedings., Eleventh Annual International Phoenix Conference on

Conference_Location

Scottsdale, AZ, USA

Print_ISBN

0-7803-0605-8

Type

conf

DOI

10.1109/PCCC.1992.200513

Filename

200513