Multiscale segmentation of unstructured document pages using soft decision integration

Kamran Etemad; David Doermann; Rama Chellappa

doi:10.1109/34.566817

Multiscale segmentation of unstructured document pages using soft decision integration

Kamran Etemad
, David Doermann
, Rama Chellappa

Department of Computer Science and Engineering

University of Maryland, College Park

Research output: Contribution to journal › Article › peer-review

71 Scopus citations

Abstract

We present an algorithm for layout-independent document page segmentation based on document texture using multiscale feature vectors and fuzzy local decision information. Multiscale feature vectors are classified locally using a neural network to allow soft/fuzzy multi-class membership assignments. Segmentation is performed by integrating soft local decision vectors to reduce their.

Original language	English
Pages (from-to)	92-95
Number of pages	4
Journal	IEEE Transactions on Pattern Analysis and Machine Intelligence
Volume	19
Issue number	1
DOIs	https://doi.org/10.1109/34.566817
State	Published - 1997

Keywords

Context dependent classification
Document processing
Multiscale analysis
Neural networks
Soft decision integration
Wavelet packets

Access to Document

10.1109/34.566817

Cite this

@article{19c9707c0d6b488084025db526dcfe69,

title = "Multiscale segmentation of unstructured document pages using soft decision integration",

abstract = "We present an algorithm for layout-independent document page segmentation based on document texture using multiscale feature vectors and fuzzy local decision information. Multiscale feature vectors are classified locally using a neural network to allow soft/fuzzy multi-class membership assignments. Segmentation is performed by integrating soft local decision vectors to reduce their.",

keywords = "Context dependent classification, Document processing, Multiscale analysis, Neural networks, Soft decision integration, Wavelet packets",

author = "Kamran Etemad and David Doermann and Rama Chellappa",

year = "1997",

doi = "10.1109/34.566817",

language = "English",

volume = "19",

pages = "92--95",

journal = "IEEE Transactions on Pattern Analysis and Machine Intelligence",

issn = "0162-8828",

publisher = "IEEE Computer Society",

number = "1",

}

TY - JOUR

T1 - Multiscale segmentation of unstructured document pages using soft decision integration

AU - Etemad, Kamran

AU - Doermann, David

AU - Chellappa, Rama

PY - 1997

Y1 - 1997

N2 - We present an algorithm for layout-independent document page segmentation based on document texture using multiscale feature vectors and fuzzy local decision information. Multiscale feature vectors are classified locally using a neural network to allow soft/fuzzy multi-class membership assignments. Segmentation is performed by integrating soft local decision vectors to reduce their.

AB - We present an algorithm for layout-independent document page segmentation based on document texture using multiscale feature vectors and fuzzy local decision information. Multiscale feature vectors are classified locally using a neural network to allow soft/fuzzy multi-class membership assignments. Segmentation is performed by integrating soft local decision vectors to reduce their.

KW - Context dependent classification

KW - Document processing

KW - Multiscale analysis

KW - Neural networks

KW - Soft decision integration

KW - Wavelet packets

UR - https://www.scopus.com/pages/publications/0030871179

U2 - 10.1109/34.566817

DO - 10.1109/34.566817

M3 - Article

AN - SCOPUS:0030871179

SN - 0162-8828

VL - 19

SP - 92

EP - 95

JO - IEEE Transactions on Pattern Analysis and Machine Intelligence

JF - IEEE Transactions on Pattern Analysis and Machine Intelligence

IS - 1

ER -

Multiscale segmentation of unstructured document pages using soft decision integration

Abstract

Keywords

Access to Document

Other files and links

Fingerprint

Cite this