A class-modular feedforward neural network for handwriting recognition

Pattern Recognition

Volumn 35, Issue 1, 2002, Pages 229-244

  Oh, Il Seok ^a   Suen, Ching Y ^b  

^a Chonbuk National University   (South Korea)

^b Gina Cody School of Engineering and Computer Science   (Canada)

Author keywords

Character recognition; Class modularity; Feedforward neural network; Large set classification

Indexed keywords

DECISION THEORY; FEEDFORWARD NEURAL NETWORKS; OPTIMIZATION;

HANDWRITING RECOGNITION;

CHARACTER RECOGNITION;

ARTIFICIAL NEURAL NETWORK; IMAGE CLASSIFICATION; PATTERN RECOGNITION;

EID: 0036132148     PISSN: 00313203     EISSN: None     Source Type: Journal    
DOI: 10.1016/S0031-3203(00)00181-3     Document Type: Article

References (37)

1. Learning internal representations by error propagation
2. Handwritten Zip code recognition with multilayer networks
3. Computer recognition of unconstrained handwritten numerals
4. Gradient-based contour encoding for character recognition
5. Neural-network classifiers for recognizing totally unconstrained handwritten numerals
6. Distance features for neural network-based recognition of handwritten characters
7. Handwriting recognition or reading? Situation at the dawn of the 3rd millennium
8. A self-organizing neural tree for large-set pattern classification
9. A preclassification method for handwritten Chinese character recognition via fuzzy rules and SEART neural net
10. Sorting and recognizing cheques and financial documents
11. Strictly local backpropagation
12. Bounds on the number of samples needed for neural learning
13. Pruning algorithms-a survey
14. A hierarchical neural network architecture for handwritten numeral recognition
15. Statistical estimation of the number of hidden units for feedforward neural networks
16. Dimensionality and sample size considerations in pattern recognition practice
17. Handwritten digits recognition system via OCON neural network by pruning selective update
18. Anadaptive modular neural network with application to unconstrained character recognition
19. Efficient classification for multiclass problems using modular neural networks
20. Class-expert approach to handwritten numeral recognition
21. Pipelining multiple algorithms for handwritten numeral recognition
22. A competitive modular connectionist architecture
23. Modular learning in hierarchical neural networks
24. Modular learning
25. Modular recurrent neural networks for Mandarin syllable recognition
26. Modularity and scaling in large phonemic neural networks
27. The design of a neural network with a biologically motivated architecture
28. An improved algorithm for neural network classification of imbalanced training sets
29. A segmentation-free recognition of two touching numerals using neural networks
30. Handwritten Korean character image database PE92
31. Lexicon driven approach for offline recognition of unconstrained handwritten Korean words