A maximum-likelihood approach to segmentation-based recognition of unconstrained handwriting text

Author

Senda, Shuji ; Yamada, Keiji

Author_Institution

Comput. & Commun. Media Res., NEC Corp., Nara, Japan

fYear

2001

fDate

6/23/1905 12:00:00 AM

Firstpage

184

Lastpage

188

Abstract

We propose a maximum-likelihood approach to segmentation-based recognition of unconstrained handwriting text. The segmentation scores and recognition scores are transformed into posterior probabilities, and the likelihood function which is composed of both these probabilities and character n-gram probabilities is derived from the Bayesian theorem. The recognition result which maximizes the function can be obtained by Viterbi search. Experiments have shown that the proposed likelihood function is effective in the recognition of online Japanese text

Keywords

Bayes methods; document image processing; handwritten character recognition; image segmentation; maximum likelihood estimation; optical character recognition; probability; Bayesian theorem; OCR; Viterbi search; character n-gram probabilities; experiments; maximum-likelihood approach; online Japanese text; posterior probabilities; segmentation-based recognition; text recognition; unconstrained handwriting recognition; Bayesian methods; Character generation; Character recognition; Dictionaries; Flowcharts; Handwriting recognition; Lattices; National electric code; Text recognition; Viterbi algorithm;

fLanguage

English

Publisher

ieee

Conference_Titel

Document Analysis and Recognition, 2001. Proceedings. Sixth International Conference on

Conference_Location

Seattle, WA

Print_ISBN

0-7695-1263-1

Type

conf

DOI

10.1109/ICDAR.2001.953780

Filename

953780