Title :
On-line handwriting recognition with a neuro-fuzzy method
Author_Institution :
ATR Human Inf. Process. Res. Labs., Kyoto, Japan
Abstract :
This paper describes an efficient neuro-fuzzy method for recognizing online handwriting characters. The basic idea is to train a number of network classifiers and aggregating them with fuzzy logic. The method combines the outputs of separate networks with importance of each network, which is subjectively assigned as the nature of fuzzy logic. We demonstrate the superior performance of the presented method and compare with conventional methods like voting and averaging by thorough experiments on a difficult online handwriting recognition problem
Keywords :
character recognition; fuzzy neural nets; learning (artificial intelligence); aggregation; character recognition; efficient neuro-fuzzy method; fuzzy logic; network classifiers; online handwriting recognition; Character recognition; Filtering; Fuzzy logic; Handwriting recognition; Hardware; Humans; Information processing; Neural networks; Smoothing methods; Writing;
Conference_Titel :
Fuzzy Systems, 1995. International Joint Conference of the Fourth IEEE International Conference on Fuzzy Systems and The Second International Fuzzy Engineering Symposium., Proceedings of 1995 IEEE Int
Conference_Location :
Yokohama
Print_ISBN :
0-7803-2461-7
DOI :
10.1109/FUZZY.1995.409825