Title :
High performance OCR with syntactic neural networks
Author_Institution :
Essex Univ., Colchester, UK
Abstract :
This paper describes the application of a special type of syntactic neural network (SNN) to the recognition of hand-written digits. Importantly, it is shown that this class of SNN can be implemented to work at very high classification speeds (similar to that of an N-tuple classifier), but with higher classification accuracy when trained on enough data. Results are reported on the ESSEX and CEDAR data sets to demonstrate this
Keywords :
image classification; neural nets; optical character recognition; CEDAR data set; ESSEX data set; N-tuple classifiers; classification accuracy; classification speed; handwritten digit recognition; high-performance OCR; optical character recognition; syntactic neural networks; training;
Conference_Titel :
Artificial Neural Networks, 1995., Fourth International Conference on
Conference_Location :
Cambridge
Print_ISBN :
0-85296-641-5
DOI :
10.1049/cp:19950542
