Title :
Distortion invariant handwritten digit recognition using adaptive resonance theory (ART) neural net model
Author :
Khan, Emdadur R.
Author_Institution :
Nat. Semicond., Santa Clara, CA, USA
Abstract :
A distortion- and rotation-invariant handwritten digit recognition scheme using a modified version of the adaptive resonance theory neural network model proposed by S. Grossberg and G. Carpenter (1988) is reported. The scheme is robust. It can be extended to the recognition of other handwritten characters
Keywords :
adaptive systems; neural nets; optical character recognition; adaptive resonance theory neural network model; distortion-invariant recognition; handwritten characters; rotation-invariant handwritten digit recognition; Character recognition; Handwriting recognition; Neural networks; Neurons; Reliability theory; Resonance; Robustness; Shape; Subspace constraints; Very large scale integration;
Conference_Titel :
Industrial Electronics Society, 1990. IECON '90., 16th Annual Conference of IEEE
Conference_Location :
Pacific Grove, CA
Print_ISBN :
0-87942-600-4
DOI :
10.1109/IECON.1990.149319
