Massively-parallel handwritten character recognition based on the distance transform

Pattern Recognition

Volumn 28, Issue 3, 1995, Pages 293-301

  Kovacs V ZS M ^a   Guerrieri, R ^a  

^a UNIVERSITY OF BOLOGNA   (Italy)

Author keywords

Classifiers; Connection Machine; Digits; Distance transform; Massively parallel processing; Nearest neighbors; OCR; Pattern recognition; Template matching; Upper case letters

Indexed keywords

CLASSIFICATION (OF INFORMATION); COMPUTATIONAL GEOMETRY; DATABASE SYSTEMS; IMAGE PROCESSING; MATHEMATICAL TRANSFORMATIONS; PARALLEL PROCESSING SYSTEMS; STATISTICAL METHODS;

ACCURACY; BINARIZATION NORMALIZATION; CONNECTION MACHINE; DISTANCE TRANSFORM; GRAY SCALE PICTURE; HANDWRITTEN CHARACTER RECOGNITION; IMAGE BINARIZATION; STATISTICAL CLASSIFIER; UPPER SCALE LETTERS;

CHARACTER RECOGNITION;

EID: 0029269679     PISSN: 00313203     EISSN: None     Source Type: Journal    
DOI: 10.1016/0031-3203(94)00099-8     Document Type: Article

References (24)

1. Structural classification and relaxation matching of totally unconstrained handwritten zip-code numbers
2. Recognition of handwritten digits using template and model matching
3. Handwritten digit recognition with a back-propagation network
4. A neural net OCR using geometrical and zonal pattern features
5. A new approach to handwritten character recognition
6. Sequential operations in digital picture processing
7. Distance transformations in digital images
8. A note on binary template matching
9. Moment normalization of handprinted characters
10. Parallel thinning with twosubiteration algorithms
11. Distance transforms: properties and machine vision applications
12. The nearest neighbor classification rule with reject option
13. Partial shape classification using contour matching in distance transformation
14. Massively parallel AI
15. Memory-based reasoning approach for pattern recognition of binary images
16. Small sample size effects in statistical pattern recognition: recommendations for practitioners
17. A note on geometric transforms' of digital sets
18. Nistir 4912