DocumentCode
3254272
Title
Fitting elastic maps to recognize handwritten digits
Author
Lim, J.H. ; Teh, H.H. ; Lui, H.C. ; Wang, P.Z.
Author_Institution
Neuro ISS Lab., RWCP, Kent Ridge, Singapore
Volume
6
fYear
1995
fDate
Nov/Dec 1995
Firstpage
3078
Abstract
In this paper, we present a novel approach to recognize off-line handwritten characters by fitting elastic topological maps. A supervised incremental clustering algorithm is designed to learn stochastic prototypes from examples with elastic matching. We report experimental results on NIST SD3 digit recognition using our proposed approach and draw a connection to deformable models
Keywords
character recognition; feature extraction; fuzzy set theory; image matching; learning (artificial intelligence); self-organising feature maps; topology; NIST SD3 digit recognition; deformable models; elastic matching; elastic topological maps; feature extraction; fitting elastic maps; fuzzy c-means; neural networks; off-line handwritten character recognition; supervised incremental clustering; Character recognition; Clustering algorithms; Deformable models; Feature extraction; Handwriting recognition; Laboratories; Network topology; Neural networks; Prototypes; Stochastic processes;
fLanguage
English
Publisher
ieee
Conference_Titel
Neural Networks, 1995. Proceedings., IEEE International Conference on
Conference_Location
Perth, WA
Print_ISBN
0-7803-2768-3
Type
conf
DOI
10.1109/ICNN.1995.487275
Filename
487275
Link To Document