Neural network-based systems for handprint OCR applications
- 1 January 1998
- journal article
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Image Processing
- Vol. 7 (8) , 1097-1112
- https://doi.org/10.1109/83.704304
Abstract
Over the last five years or so, neural network (NN)-based approaches have been steadily gaining performance and popularity for a wide range of optical character recognition (OCR) problems, from isolated digit recognition to handprint recognition. We present an NN classification scheme based on an enhanced multilayer perceptron (MLP) and describe an end-to-end system for form-based handprint OCR applications designed by the National Institute of Standards and Technology (NIST) Visual Image Processing Group. The enhancements to the MLP are based on (i) neuron activations functions that reduce the occurrences of singular Jacobians; (ii) successive regularization to constrain the volume of the weight space; and (iii) Boltzmann pruning to constrain the dimension of the weight space. Performance characterization studies of NN systems evaluated at the first OCR systems conference and the NIST form-based handprint recognition system are also summarized.Keywords
This publication has 37 references indexed in Scilit:
- Large vocabulary recognition of on-line handwritten cursive wordsPublished by Institute of Electrical and Electronics Engineers (IEEE) ,1996
- Neural network positioning and classification of handwritten charactersNeural Networks, 1996
- Feature extraction methods for character recognition-A surveyPattern Recognition, 1996
- Binary decision clustering for neural-network-based optical character recognitionPattern Recognition, 1996
- Regularization Theory and Neural Networks ArchitecturesNeural Computation, 1995
- Evaluation of pattern classifiers for fingerprint and OCR applicationsPattern Recognition, 1994
- Information Content in Neural Net OptimizationConnection Science, 1994
- Centered-Object Integrated Segmentation and Recognition of Overlapping Handprinted CharactersNeural Computation, 1993
- Recognizing Hand-Printed Letters and Digits Using Backpropagation LearningNeural Computation, 1991
- Properties of center manifoldsTransactions of the American Mathematical Society, 1985