This paper demonstrates effectiveness of training of Fuzzy Spline Curve Identifier (FSCI) using a fuzzy neural network. FSCI was proposed as a primitive curve identification system designed to establish a general-purpose freehand interface for computer aided drawing (CAD) systems. It succeeded in distinguishing a freehand drawing into seven kinds of primitive curves which are indispensable for use in CAD. The key was the introduction of a fuzzy reasoning which embodied a strategy to try to find the simplest primitive curves in drawing. A trainable version of FSCI was then proposed, by introducing a structured fuzzy neural network, in order that it would acquire learning ability to adapt itself to individual drawing manner. This paper sets up some experiment on FSCI and demonstrates the effectiveness of the training by evaluating curve class recognition rates. Furthermore, through some considerations on a concrete example of the training, it shows that the introduced fuzzy neural network is informative for us to analyze users' drawing manner and also the identification characteristics of FSCI.
Training of the Freehand Curve Identifier FSCI Using a Fuzzy Neural Network
50_irai01.pdf
(539.35KB)
