A trainable freehand curve identifier using a fuzzy neural network

The authors have proposed the Fuzzy Spline Curve Identification (FSCI) method, which is a basic technology of general‐purpose freehand‐curve input for a CAD system. FSCI identifies seven curve primitives which are basic elements of CAD patterns from freehand drawings by assuming the simplest curve p...

Full description

Saved in:
Bibliographic Details
Published in:Electronics & communications in Japan. Part 2, Electronics Electronics, 2002-10, Vol.85 (10), p.51-59
Main Authors: Mori, Saori, Saga, Sato
Format: Article
Language:English
Subjects:
freehand drawing
fuzzy system
neural network
optimization
pattern recognition
Citations: Items that this one cites
Online Access:Get full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:The authors have proposed the Fuzzy Spline Curve Identification (FSCI) method, which is a basic technology of general‐purpose freehand‐curve input for a CAD system. FSCI identifies seven curve primitives which are basic elements of CAD patterns from freehand drawings by assuming the simplest curve primitives in ambiguous freehand curves. However, this method requires considerable user training in drawing due to the fixed fuzzy rules used in FSCI. This paper proposes a new learning‐type FSCI in which the rules are flexible so that user training can be reduced. The method is a combination of fuzzy theory and the learning capability of a neural network. It realizes a learning‐type FSCI which matches the type of each user drawing without losing the features of conventional FSCI. © 2002 Wiley Periodicals, Inc. Electron Comm Jpn Pt 2, 85(10): 51–59, 2002; Published online in Wiley InterScience (www.interscience.wiley.com). DOI 10.1002/ecjb.10089
ISSN:8756-663X
1520-6432
DOI:10.1002/ecjb.10089