Presentation 1993/6/18
Shape recognition using adaptive VQ and neural networks
Masaji Katagiri, Masakazu Nagura, Hiroyuki Arai,
PDF Download Page PDF download Page Link
Abstract(in Japanese) (See Japanese page)
Abstract(in English) Artificial neural networks and adaptive vector quantization are applied to the shape recognition problem for closed polylines.Here polylines consist of not only straight lines but curvs. Curvature is employed to capture figures.This method makes possible to recognize, classify figures without embedding domain specific algorithms.Neural networks provide learning facility.As preprocessing of the neural networks,curvs are quantized adaptively.The quantized data and scale factors are given to the neural networks.The system can recognize figures irrelevantly to position,rotation and scaling.Some experimental results are shown.
Keyword(in Japanese) (See Japanese page)
Keyword(in English) recognition / artificial neural networks / adaptive vector quantization / learning / classification
Paper # PRU93-19
Date of Issue

Conference Information
Committee PRU
Conference Date 1993/6/18(1days)
Place (in Japanese) (See Japanese page)
Place (in English)
Topics (in Japanese) (See Japanese page)
Topics (in English)
Chair
Vice Chair
Secretary
Assistant

Paper Information
Registration To Pattern Recognition and Understanding (PRU)
Language JPN
Title (in Japanese) (See Japanese page)
Sub Title (in Japanese) (See Japanese page)
Title (in English) Shape recognition using adaptive VQ and neural networks
Sub Title (in English)
Keyword(1) recognition
Keyword(2) artificial neural networks
Keyword(3) adaptive vector quantization
Keyword(4) learning
Keyword(5) classification
1st Author's Name Masaji Katagiri
1st Author's Affiliation NTT Human Interface Laboratories()
2nd Author's Name Masakazu Nagura
2nd Author's Affiliation NTT Human Interface Laboratories
3rd Author's Name Hiroyuki Arai
3rd Author's Affiliation NTT Human Interface Laboratories
Date 1993/6/18
Paper # PRU93-19
Volume (vol) vol.93
Number (no) 100
Page pp.pp.-
#Pages 8
Date of Issue