Content features for logical document labeling

Jian Liang; David Doermann

doi:10.1117/12.476061

Content features for logical document labeling

Jian Liang
, David Doermann

Department of Computer Science and Engineering

University of Maryland, College Park

Research output: Contribution to journal › Conference article › peer-review

7 Scopus citations

Abstract

The use of content features extracted from recognized text is valuable in labeling logical elements in documents without rigid layout structure, like business letters. This paper discusses a model-based approach to combining content features with other geometrical and presentation features for logical labeling. Models are automatically initialized and adaptively improved using training samples. Satisfactory experimental results are presented.

Original language	English
Pages (from-to)	189-196
Number of pages	8
Journal	Proceedings of SPIE - The International Society for Optical Engineering
Volume	5010
DOIs	https://doi.org/10.1117/12.476061
State	Published - 2003
Event	Document Recognition and Retrieval X - Santa Clara, CA, United States Duration: Jan 22 2003 → Jan 24 2003

Keywords

Content feature
Logical document labeling

Access to Document

10.1117/12.476061

Cite this

@article{b83c0626838649ffb90992d2fa5853d1,

title = "Content features for logical document labeling",

abstract = "The use of content features extracted from recognized text is valuable in labeling logical elements in documents without rigid layout structure, like business letters. This paper discusses a model-based approach to combining content features with other geometrical and presentation features for logical labeling. Models are automatically initialized and adaptively improved using training samples. Satisfactory experimental results are presented.",

keywords = "Content feature, Logical document labeling",

author = "Jian Liang and David Doermann",

year = "2003",

doi = "10.1117/12.476061",

language = "English",

volume = "5010",

pages = "189--196",

journal = "Proceedings of SPIE - The International Society for Optical Engineering",

issn = "0277-786X",

publisher = "SPIE",

note = "Document Recognition and Retrieval X ; Conference date: 22-01-2003 Through 24-01-2003",

}

TY - JOUR

T1 - Content features for logical document labeling

AU - Liang, Jian

AU - Doermann, David

PY - 2003

Y1 - 2003

N2 - The use of content features extracted from recognized text is valuable in labeling logical elements in documents without rigid layout structure, like business letters. This paper discusses a model-based approach to combining content features with other geometrical and presentation features for logical labeling. Models are automatically initialized and adaptively improved using training samples. Satisfactory experimental results are presented.

AB - The use of content features extracted from recognized text is valuable in labeling logical elements in documents without rigid layout structure, like business letters. This paper discusses a model-based approach to combining content features with other geometrical and presentation features for logical labeling. Models are automatically initialized and adaptively improved using training samples. Satisfactory experimental results are presented.

KW - Content feature

KW - Logical document labeling

UR - https://www.scopus.com/pages/publications/0038056107

U2 - 10.1117/12.476061

DO - 10.1117/12.476061

M3 - Conference article

AN - SCOPUS:0038056107

SN - 0277-786X

VL - 5010

SP - 189

EP - 196

JO - Proceedings of SPIE - The International Society for Optical Engineering

JF - Proceedings of SPIE - The International Society for Optical Engineering

T2 - Document Recognition and Retrieval X

Y2 - 22 January 2003 through 24 January 2003

ER -

Content features for logical document labeling

Abstract

Keywords

Access to Document

Other files and links

Fingerprint

Cite this