Handwritten Character Recognition Using Gray-scale Based State-Space Parameters and Class Modular NN
Author:
- 1 January 2008
- conference paper
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- p. 374-379
Abstract
We present a novel feature extraction method for offline recognition of segmented Malayalam handwritten characters from their gray-scale images without the usual step of binarization. We investigate a new approach to model handwritten characters using state-space map (SSM) and state-space point distribution (SSPD) parameters. In the recognition stage we used class modular neural network with the proposed SSPD features and this method is found to be promising.Keywords
This publication has 9 references indexed in Scilit:
- Nonlinear Time Series Analysis2003
- Efficient classification for multiclass problems using modular neural networksIEEE Transactions on Neural Networks, 1995
- AN ADAPTIVE MODULAR NEURAL NETWORK WITH APPLICATION TO UNCONSTRAINED CHARACTER RECOGNITIONInternational Journal of Pattern Recognition and Artificial Intelligence, 1994
- Direct gray-scale extraction of features for character recognitionIEEE Transactions on Pattern Analysis and Machine Intelligence, 1993
- Chaotic Dynamics: An IntroductionPhysics Today, 1990
- On the Recognition of Printed Characters of Any Font and SizeIEEE Transactions on Pattern Analysis and Machine Intelligence, 1987
- Extracting qualitative dynamics from experimental dataPhysica D: Nonlinear Phenomena, 1986
- Detecting strange attractors in turbulence1981
- Detection of Surface-Specific Points by Local Parallel Processing of Discrete Terrain Elevation DataComputer Graphics and Image Processing, 1975