A neural network approach to discrimination between defects and calyces in oranges

Le Matematiche. 1993;48(2):273-285

 

Journal Homepage

Journal Title: Le Matematiche

ISSN: 0373-3505 (Print); 2037-5298 (Online)

Publisher: Universit√† degli Studi di Catania

Society/Institution: University of Catania

LCC Subject Category: Science: Mathematics

Country of publisher: Italy

Language of fulltext: Italian, French, English

Full-text formats available: PDF

 

AUTHORS

Salvatore Ingrassia
Enrico Commis

EDITORIAL INFORMATION

Peer review

Editorial Board

Instructions for authors

Time From Submission to Publication: 53 weeks

 

Abstract | Full Text

<span style="font-family: DejaVu Sans,sans-serif;">The problem of automatic discrimination among pictures concerning either defects or calyces in oranges is approached. The method here proposed is based on a statistical analysis of the grey-levels and the shape of calyces in the pictures. Some suitable statistical indices are considered and the discriminant function is designed by means of a neural network on the basis of a suitable vector representation of the images. Numerical experiments give 5 misclassifications in a set of 52 images, where only three defects have been classified as calyces.</span>