DocumentCode :
1908753
Title :
SCRIPT-a prototype for the recognition of continuous, cursive, handwritten input by means of a neural network simulator
Author :
Harrienhausen-Muhlbauer, B. ; Koop, Andreas
Author_Institution :
IBM Germany, Heidelberg, Germany
fYear :
1993
fDate :
1993
Firstpage :
1672
Abstract :
SCRIPT, a prototype for the recognition of continuous handwriting, where the recognition is independent from the size and resolution of the handwritten input is described. The system has presently stored (learned) different handwritings, but individual customization, i.e., learning of a new handwriting, is possible anytime. SCRIPT is a modular system, which uses several neural nets as well as a fuzzy logic component. The system is designed so that different types of input, e.g., scanned handwriting as well as input via a pen-computer, is possible. For reasons of testing, input via scanned images are chosen
Keywords :
character recognition; fuzzy logic; image scanners; neural nets; SCRIPT; character recognition; continuous handwriting; customization; fuzzy logic component; handwritten input; modular system; neural network simulator; scanned images; Biological neural networks; Clustering algorithms; Fuzzy logic; Handwriting recognition; Humans; Image storage; Pixel; Prototypes; Virtual prototyping; Workstations;
fLanguage :
English
Publisher :
ieee
Conference_Titel :
Neural Networks, 1993., IEEE International Conference on
Conference_Location :
San Francisco, CA
Print_ISBN :
0-7803-0999-5
Type :
conf
DOI :
10.1109/ICNN.1993.298808
Filename :
298808
Link To Document :
بازگشت