Modeling the sample distribution for clustering OCR

Proceedings of SPIE - The International Society for Optical Engineering

Volumn 4307, Issue , 2001, Pages 193-199

  Breuel, T M ^a  

^a PALO ALTO RESEARCH CENTER   (United States)

Author keywords

Bayesian; Font independent; Mixture of gaussians; Optical character recognition; Sample distribution

Indexed keywords

ALGORITHMS; COMPUTER SIMULATION; IMAGE COMPRESSION; PATTERN RECOGNITION; STATISTICAL METHODS;

CLUSTERING;

OPTICAL CHARACTER RECOGNITION;

EID: 0035058053     PISSN: 0277786X     EISSN: None     Source Type: Conference Proceeding    
DOI: 10.1117/12.410858     Document Type: Conference Paper

References (15)

1. Unsupervised construction of decision networks for pattern classification
2. Text ocr by solving a cryptogram
3. Efficient algorithms to decode substitution ciphers with applications to ocr
4. Digipaper: A versatile color document image representation
5. Color documents on the web with djvu
6. Enhancing degraded document images via bitmap clustering and averaging
7. Improving ocr performance with word image equivalence
8. Estimation and validation of document degradation models
9. A non-linear mapping for data structure analysis