A Comparison of Binarization Methods for Historical Archive Documents.

Jingyu He Q. D. M. Do Andy C. Downton J. H. Kim

Published in: ICDAR (2005)

Keyphrases

preprocessing
machine learning methods
significant improvement
information retrieval
metadata
benchmark datasets
text corpora
data collections
data mining
xml documents
computational cost
web search
empirical studies
document classification
vector space model
latent semantic analysis