Loading [MathJax]/jax/input/TeX/config.js

A neural approach for text extraction from scholarly figures

Publikation: Beitrag in Buch/Bericht/Sammelwerk/KonferenzbandAufsatz in KonferenzbandForschungPeer-Review

Autorschaft

  • David Morris
  • Peichen Tang
  • Ralph Ewerth

Organisationseinheiten

Externe Organisationen

  • Technische Informationsbibliothek (TIB) Leibniz-Informationszentrum Technik und Naturwissenschaften und Universitätsbibliothek
Plum Print visual indicator of research metrics
  • Citations
    • Citation Indexes: 8
  • Captures
    • Readers: 10
see details

Details

OriginalspracheEnglisch
Titel des Sammelwerks2019 International Conference on Document Analysis and Recognition (ICDAR)
UntertitelProceedings
Herausgeber (Verlag)Institute of Electrical and Electronics Engineers Inc.
Seiten1438-1443
Seitenumfang6
ISBN (elektronisch)978-1-7281-3014-9
ISBN (Print)978-1-7281-3015-6
PublikationsstatusVeröffentlicht - Sept. 2019
Veranstaltung15th IAPR International Conference on Document Analysis and Recognition, ICDAR 2019 - Sydney, Australien
Dauer: 20 Sept. 201925 Sept. 2019

Publikationsreihe

NameProceedings of the International Conference on Document Analysis and Recognition
ISSN (Print)1520-5363
ISSN (elektronisch)2379-2140

Abstract

In recent years, the problem of scene text extraction from images has received extensive attention and significant progress. However, text extraction from scholarly figures such as plots and charts remains an open problem, in part due to the difficulty of locating irregularly placed text lines. To the best of our knowledge, literature has not described the implementation of a text extraction system for scholarly figures that adapts deep convolutional neural networks used for scene text detection. In this paper, we propose a text extraction approach for scholarly figures that forgoes preprocessing in favor of using a deep convolutional neural network for text line localization. Our system uses a publicly available scene text detection approach whose network architecture is well suited to text extraction from scholarly figures. Training data are derived from charts in arXiv papers which are extracted using Allen Institute's pdffigures tool. Since this tool analyzes PDF data as a container format in order to extract text location through the mechanisms which render it, we were able to gather a large set of labeled training samples. We show significant improvement from methods in the literature, and discuss the structural changes of the text extraction pipeline.

ASJC Scopus Sachgebiete

Zitieren

A neural approach for text extraction from scholarly figures. / Morris, David; Tang, Peichen; Ewerth, Ralph.
2019 International Conference on Document Analysis and Recognition (ICDAR): Proceedings. Institute of Electrical and Electronics Engineers Inc., 2019. S. 1438-1443 8978202 (Proceedings of the International Conference on Document Analysis and Recognition).

Publikation: Beitrag in Buch/Bericht/Sammelwerk/KonferenzbandAufsatz in KonferenzbandForschungPeer-Review

Morris, D, Tang, P & Ewerth, R 2019, A neural approach for text extraction from scholarly figures. in 2019 International Conference on Document Analysis and Recognition (ICDAR): Proceedings., 8978202, Proceedings of the International Conference on Document Analysis and Recognition, Institute of Electrical and Electronics Engineers Inc., S. 1438-1443, 15th IAPR International Conference on Document Analysis and Recognition, ICDAR 2019, Sydney, Australien, 20 Sept. 2019. https://doi.org/10.1109/ICDAR.2019.00231
Morris, D., Tang, P., & Ewerth, R. (2019). A neural approach for text extraction from scholarly figures. In 2019 International Conference on Document Analysis and Recognition (ICDAR): Proceedings (S. 1438-1443). Artikel 8978202 (Proceedings of the International Conference on Document Analysis and Recognition). Institute of Electrical and Electronics Engineers Inc.. https://doi.org/10.1109/ICDAR.2019.00231
Morris D, Tang P, Ewerth R. A neural approach for text extraction from scholarly figures. in 2019 International Conference on Document Analysis and Recognition (ICDAR): Proceedings. Institute of Electrical and Electronics Engineers Inc. 2019. S. 1438-1443. 8978202. (Proceedings of the International Conference on Document Analysis and Recognition). doi: 10.1109/ICDAR.2019.00231
Morris, David ; Tang, Peichen ; Ewerth, Ralph. / A neural approach for text extraction from scholarly figures. 2019 International Conference on Document Analysis and Recognition (ICDAR): Proceedings. Institute of Electrical and Electronics Engineers Inc., 2019. S. 1438-1443 (Proceedings of the International Conference on Document Analysis and Recognition).
Download
@inproceedings{fe759be2004f4a4a9d58903e632e5e7b,
title = "A neural approach for text extraction from scholarly figures",
abstract = "In recent years, the problem of scene text extraction from images has received extensive attention and significant progress. However, text extraction from scholarly figures such as plots and charts remains an open problem, in part due to the difficulty of locating irregularly placed text lines. To the best of our knowledge, literature has not described the implementation of a text extraction system for scholarly figures that adapts deep convolutional neural networks used for scene text detection. In this paper, we propose a text extraction approach for scholarly figures that forgoes preprocessing in favor of using a deep convolutional neural network for text line localization. Our system uses a publicly available scene text detection approach whose network architecture is well suited to text extraction from scholarly figures. Training data are derived from charts in arXiv papers which are extracted using Allen Institute's pdffigures tool. Since this tool analyzes PDF data as a container format in order to extract text location through the mechanisms which render it, we were able to gather a large set of labeled training samples. We show significant improvement from methods in the literature, and discuss the structural changes of the text extraction pipeline.",
keywords = "Document analysis, Figure search, Neural networks, Text detection, Text extraction",
author = "David Morris and Peichen Tang and Ralph Ewerth",
note = "Funding information: This work is financially supported by the German Federal Ministry of Education and Research (BMBF) and European Social Fund (ESF) (InclusiveOCW project, no. 01PE17004).; 15th IAPR International Conference on Document Analysis and Recognition, ICDAR 2019 ; Conference date: 20-09-2019 Through 25-09-2019",
year = "2019",
month = sep,
doi = "10.1109/ICDAR.2019.00231",
language = "English",
isbn = "978-1-7281-3015-6",
series = "Proceedings of the International Conference on Document Analysis and Recognition",
publisher = "Institute of Electrical and Electronics Engineers Inc.",
pages = "1438--1443",
booktitle = "2019 International Conference on Document Analysis and Recognition (ICDAR)",
address = "United States",

}

Download

TY - GEN

T1 - A neural approach for text extraction from scholarly figures

AU - Morris, David

AU - Tang, Peichen

AU - Ewerth, Ralph

N1 - Funding information: This work is financially supported by the German Federal Ministry of Education and Research (BMBF) and European Social Fund (ESF) (InclusiveOCW project, no. 01PE17004).

PY - 2019/9

Y1 - 2019/9

N2 - In recent years, the problem of scene text extraction from images has received extensive attention and significant progress. However, text extraction from scholarly figures such as plots and charts remains an open problem, in part due to the difficulty of locating irregularly placed text lines. To the best of our knowledge, literature has not described the implementation of a text extraction system for scholarly figures that adapts deep convolutional neural networks used for scene text detection. In this paper, we propose a text extraction approach for scholarly figures that forgoes preprocessing in favor of using a deep convolutional neural network for text line localization. Our system uses a publicly available scene text detection approach whose network architecture is well suited to text extraction from scholarly figures. Training data are derived from charts in arXiv papers which are extracted using Allen Institute's pdffigures tool. Since this tool analyzes PDF data as a container format in order to extract text location through the mechanisms which render it, we were able to gather a large set of labeled training samples. We show significant improvement from methods in the literature, and discuss the structural changes of the text extraction pipeline.

AB - In recent years, the problem of scene text extraction from images has received extensive attention and significant progress. However, text extraction from scholarly figures such as plots and charts remains an open problem, in part due to the difficulty of locating irregularly placed text lines. To the best of our knowledge, literature has not described the implementation of a text extraction system for scholarly figures that adapts deep convolutional neural networks used for scene text detection. In this paper, we propose a text extraction approach for scholarly figures that forgoes preprocessing in favor of using a deep convolutional neural network for text line localization. Our system uses a publicly available scene text detection approach whose network architecture is well suited to text extraction from scholarly figures. Training data are derived from charts in arXiv papers which are extracted using Allen Institute's pdffigures tool. Since this tool analyzes PDF data as a container format in order to extract text location through the mechanisms which render it, we were able to gather a large set of labeled training samples. We show significant improvement from methods in the literature, and discuss the structural changes of the text extraction pipeline.

KW - Document analysis

KW - Figure search

KW - Neural networks

KW - Text detection

KW - Text extraction

UR - http://www.scopus.com/inward/record.url?scp=85079880203&partnerID=8YFLogxK

U2 - 10.1109/ICDAR.2019.00231

DO - 10.1109/ICDAR.2019.00231

M3 - Conference contribution

AN - SCOPUS:85079880203

SN - 978-1-7281-3015-6

T3 - Proceedings of the International Conference on Document Analysis and Recognition

SP - 1438

EP - 1443

BT - 2019 International Conference on Document Analysis and Recognition (ICDAR)

PB - Institute of Electrical and Electronics Engineers Inc.

T2 - 15th IAPR International Conference on Document Analysis and Recognition, ICDAR 2019

Y2 - 20 September 2019 through 25 September 2019

ER -