Neural network techniques for fine-form discrimination

Author

Germagnoli, Fabio ; Lazzari, Stefano ; Magenes, Giovanni

Author_Institution

Dipt. di Inf. e Sistemistica, Pavia Univ., Italy

Volume

1

fYear

1995

fDate

20-25 Sep 1995

Firstpage

811

Abstract

The ability of neural network (NN) techniques in processing tensorial tactile images have been investigated. Tactile images were provided by a sensor array of 21×21 sensitive elements, realized with piezoelectric polymers, able to detect the six components of the stress tensor. Two different types of NN have been used: a multilayer perceptron trained with the backpropagation algorithm, to filter and pre-process the data (reconstruction phase) and an ART network, to recognize different object typology (classification phase)

Keywords

ART neural nets; array signal processing; backpropagation; image classification; image reconstruction; multilayer perceptrons; piezoelectric transducers; stress measurement; tactile sensors; 21 by 21 elements; 21×21 sensitive elements; ART network; artificial tactile system; backpropagation algorithm; classification phase; fine-form discrimination; multilayer perceptron; neural network techniques; object typology; piezoelectric polymers; reconstruction phase; robots; sensor array; stress tensor; tensorial tactile images; unstructured environments; Backpropagation algorithms; Filters; Image reconstruction; Image sensors; Multilayer perceptrons; Neural networks; Polymers; Sensor arrays; Tactile sensors; Tensile stress;

fLanguage

English

Publisher

ieee

Conference_Titel

Engineering in Medicine and Biology Society, 1995., IEEE 17th Annual Conference

Conference_Location

Montreal, Que.

Print_ISBN

0-7803-2475-7

Type

conf

DOI

10.1109/IEMBS.1995.575375

Filename

575375