research-article

Document Image Binarization using LSTM: A Sequence Learning Approach

Authors:

Muhammad Zeshan Afzal,

Joan Pastor-Pellicer,

Faisal Shafait,

Thomas M. Breuel,

Andreas Dengel,

Marcus LiwickiAuthors Info & Claims

HIP '15: Proceedings of the 3rd International Workshop on Historical Document Imaging and Processing

Pages 79 - 84

https://doi.org/10.1145/2809544.2809561

Published: 22 August 2015 Publication History

Abstract

We propose to address the problem of Document Image Binarization (DIB) using Long Short-Term Memory (LSTM) which is specialized in processing very long sequences. Thus, the image is considered as a 2D sequence of pixels and in accordance to this a 2D LSTM is employed for the classification of each pixel as text or background. The proposed approach processes the information using local context and then propagates the information globally in order to achieve better visual coherence. The method is robust against most of the document artifacts. We show that with a very simple network without any feature extraction and with limited amount of data the proposed approach works reasonably well for the DIBCO 2013 dataset. Furthermore a synthetic dataset is considered to measure the performance of the proposed approach with both binarization and OCR groundtruth. The proposed approach significantly outperforms standard binarization approaches both for F-Measure and OCR accuracy with the availability of enough training samples.

References

[1]

M. Z. Afzal, M. Krämer, S. S. Bukhari, M. R. Yousefi, F. Shafait, and T. M. Breuel. Robust binarization of stereo and monocular document images using percentile filter. In C CBDAR 2013, Washington, DC, USA, August 23, 2013, pages 139--149, 2013.

[2]

N. Babaguchi and K. Yamada. Connectionist model binarization. In Pattern Recognition, pages 51--56, 1990.

[3]

E. Badekas, N. A. Nikolaou, and N. Papamarkos. Text localization and binarization in complex color documents. In MLDM Posters, pages 1--15, 2007.

[4]

R. Chamchong and C. Fung. Optimal selection of binarization techniques for the processing of ancient palm leaf manuscripts. Systems Man and Cybernetics (SMC), pages 3796--3800, 2010.

[5]

C.-H. Chou, W.-H. Lin, and F. Chang. A binarization method with learning-built rules for document images produced by cameras. Pattern Recognition, 43(4):1518--1530, Apr. 2010.

Digital Library

[6]

B. Gatos, K. Ntirogiannis, and I. Pratikakis. DIBCO 2009: document image binarization contest. IJDAR, 14(1):35--44, 2011.

Digital Library

[7]

A. Graves, M. Liwicki, S. Fernández, R. Bertolami, H. Bunke, and J. Schmidhuber. A novel connectionist system for unconstrained handwriting recognition. IEEE transactions on pattern analysis and machine intelligence, 31(5):855--68, May 2009.

[8]

A. Graves and J. Schmidhuber. Framewise phoneme classification with bidirectional LSTM and other neural network architectures. Neural networks: the official journal of the International Neural Network Society, 18(5--6):602--10, 2005.

[9]

A. Graves and J. Schmidhuber. Offline handwriting recognition with multidimensional recurrent neural networks. In Advances in Neural Information Processing Systems, pages 545--552, 2009.

Digital Library

[10]

J. Kuk, N. Cho, and K. Lee. MAP-MRF approach for binarization of degraded document image. Image Processing, 2008. ICIP 2008, pages 2612--2615, 2008.

[11]

T. Lelore and F. Bouchara. Document Image Binarisation Using Markov Field Model. In 2009 10th International Conference on Document Analysis and Recognition, number 2, pages 551--555. Ieee, 2009.

Digital Library

[12]

G. Nagy. Twenty years of document image analysis in pami. IEEE Trans. Pattern Anal. Mach. Intell., 22(1):38--62, Jan. 2000.

Digital Library

[13]

W. Niblack. An Introduction to digital image processing. Prentice-Hall, Englewood Cliffs, NJ, 1986.

Digital Library

[14]

K. Ntirogiannis, B. Gatos, and I. Pratikakis. A combined approach for the binarization of handwritten document images. Pattern Recognition Letters, 35(0):3--15, 2014. Frontiers in Handwriting Processing.

Digital Library

[15]

H. Orii, H. Kawano, H. Maeda, and N. Ikoma. Text-color-independent binarization for degraded document image based on map-mrf approach. IEICE Transactions, 94-A(11):2342--2349, 2011.

[16]

N. Otsu. A threshold selection method from gray-level histograms. Automatica, C(1):62--66, 1975.

[17]

I. Pratikakis, B. Gatos, and K. Ntirogiannis. H-DIBCO 2010 - handwritten document image binarization competition. In ICFHR 2010, Kolkata, India, 16--18 November 2010, pages 727--732, 2010.

Digital Library

[18]

I. Pratikakis, B. Gatos, and K. Ntirogiannis. Icdar 2011 document image binarization contest (dibco 2011). In Document Analysis and Recognition (ICDAR), 2011 International Conference on, pages 1506 --1510, sept. 2011.

Digital Library

[19]

I. Pratikakis, B. Gatos, and K. Ntirogiannis. ICFHR 2012 competition on handwritten document image binarization (h-dibco 2012). In ICFHR, pages 817--822, 2012.

Digital Library

[20]

I. Pratikakis, B. Gatos, and K. Ntirogiannis. ICDAR 2013 document image binarization contest (dibco 2013). In ICDAR, pages 1471--1476, 2013.

Digital Library

[21]

T. Sari, A. Kefali, and H. Bahi. An MLP for binarizing images of old manuscripts. Frontiers in Handwriting Recognition, pages 247--251, 2012.

Digital Library

[22]

J. Sauvola and M. Pietikäinen. Adaptive Document Image Binarization. Pattern Recognition, 33:225--236, 2000.

[23]

J. Schmidhuber, D. Wierstra, and F. J. Gomez. Evolino: Hybrid Neuroevolution / Optimal Linear Search for Sequence Learning Recurrent Neural Network. In 19th International Joint Conference on Artificial Intelligence (IJCAI), pages 853--858, 2005.

[24]

H. Tanaka. Threshold Correction of Document Image Binarization for Ruled-line Extraction. 2009 10th International Conference on Document Analysis and Recognition, pages 541--545, 2009.

[25]

O. Trier and A. Jain. Goal-directed evaluation of binarization methods. Pattern Analysis and Machine Intelligence, ..., 17(12):1191--1201, 1995.

[26]

C.-M. Tsai and H.-J. Lee. Binarization of color document images via luminance and saturation color features. IEEE Transactions on Image Processing, 11(4):434--451, 2002.

Digital Library

[27]

S. Wu and A. Amin. Automatic thresholding of gray-level using multistage approach. In Document Analysis and Recognition, 2003. Proceedings. Seventh International Conference on, pages 493--497 vol.1, 2003.

Digital Library

[28]

F. Yang, Z. Ma, and M. Xie. A novel binarization approach for license plate. 2006 1ST IEEE Conference on Industrial Electronics and Applications, pages 1--4, May 2006.

Cited By

Wu ZWu ZHan WYang RHe LPi H(2024)Research and application of methods to prevent external force damage to underground cables based on the IoTProceedings of the 2024 International Conference on Power Electronics and Artificial Intelligence10.1145/3674225.3674244(99-103)Online publication date: 19-Jan-2024
https://dl.acm.org/doi/10.1145/3674225.3674244
Poudel URegmi AStamenkovic ZRaja S(2024)Applicability of OCR Engines for Text Recognition in Vehicle Number Plates, Receipts and HandwritingJournal of Circuits, Systems and Computers10.1142/S021812662350321832:18Online publication date: 2-Feb-2024
https://doi.org/10.1142/S0218126623503218
Ho JLiu M(2024)Research on Document Image Binarization: A Survey2024 IEEE 7th International Conference on Electronic Information and Communication Technology (ICEICT)10.1109/ICEICT61637.2024.10671186(457-462)Online publication date: 31-Jul-2024
https://doi.org/10.1109/ICEICT61637.2024.10671186
Show More Cited By

Document Image Binarization using LSTM: A Sequence Learning Approach
1. Computing methodologies
  1. Artificial intelligence

Recommendations

Binarization of historical document images using the local maximum and minimum
DAS '10: Proceedings of the 9th IAPR International Workshop on Document Analysis Systems

This paper presents a new document image binarization technique that segments the text from badly degraded historical document images. The proposed technique makes use of the image contrast that is defined by the local image maximum and minimum. ...
Combination of Document Image Binarization Techniques
ICDAR '11: Proceedings of the 2011 International Conference on Document Analysis and Recognition

Document image binarization has been studied for decades, and many practical binarization techniques have been proposed for different kinds of document images. However, many state-of-the-art methods are particularly suitable for the document images that ...
Two-stage generative adversarial networks for binarization of color document images
Highlights
- We propose a two-stage color document image enhancement and binarization method using generative adversarial networks.
Abstract
Document image enhancement and binarization methods are often used to improve the accuracy and efficiency of document image analysis tasks such as text recognition. Traditional non-machine-learning methods are constructed on low-level ...

Comments

Information & Contributors

Information

Published In

cover image ACM Other conferences

HIP '15: Proceedings of the 3rd International Workshop on Historical Document Imaging and Processing

August 2015

155 pages

ISBN:9781450336024

DOI:10.1145/2809544

Editors:
Bertrand Coüasnon
Irisa / Insa de Rennes, France
,
Volker Märgner
Technische Universität Braunschweig, Germany
,
Volkmar Frinken
Electrical and Computer Engineering, UC Davis, USA and ONU Technology, San Jose, USA
,
Bill Barrett
Brigham Young University, USA
,
Program Chairs:
Bertrand Coüasnon
Irisa / Insa de Rennes, France
,
Volkmar Frinken
Electrical and Computer Engineering, UC Davis, USA and ONU Technology, San Jose, USA
,
Hamid Amiri
ENIT, Tunisia
,
Volker Märgner
Technische Universität Braunschweig, Germany
,
Bill Barrett
Brigham Young University, USA

Copyright © 2015 ACM.

Permission to make digital or hard copies of all or part of this work for personal or classroom use is granted without fee provided that copies are not made or distributed for profit or commercial advantage and that copies bear this notice and the full citation on the first page. Copyrights for components of this work owned by others than ACM must be honored. Abstracting with credit is permitted. To copy otherwise, or republish, to post on servers or to redistribute to lists, requires prior specific permission and/or a fee. Request permissions from [email protected]

In-Cooperation

FamilySearch: FamilySearch

Publisher

Association for Computing Machinery

New York, NY, United States

Publication History

Published: 22 August 2015

Permissions

Request permissions for this article.

Request Permissions

Check for updates

Author Tags

Qualifiers

Research-article
Research
Refereed limited

Conference

HIP '15

HIP '15: 3rd International Workshop on Historical Document Imaging and Processing

August 22, 2015

Gammarth, Tunisia

Acceptance Rates

Overall Acceptance Rate 52 of 90 submissions, 58%

Contributors

Other Metrics

View Article Metrics

Bibliometrics & Citations

Bibliometrics

Article Metrics

40
Total Citations
View Citations
345
Total Downloads

Downloads (Last 12 months)10
Downloads (Last 6 weeks)1

Reflects downloads up to 28 Feb 2025

Other Metrics

View Author Metrics

Citations

Cited By

Wu ZWu ZHan WYang RHe LPi H(2024)Research and application of methods to prevent external force damage to underground cables based on the IoTProceedings of the 2024 International Conference on Power Electronics and Artificial Intelligence10.1145/3674225.3674244(99-103)Online publication date: 19-Jan-2024
https://dl.acm.org/doi/10.1145/3674225.3674244
Poudel URegmi AStamenkovic ZRaja S(2024)Applicability of OCR Engines for Text Recognition in Vehicle Number Plates, Receipts and HandwritingJournal of Circuits, Systems and Computers10.1142/S021812662350321832:18Online publication date: 2-Feb-2024
https://doi.org/10.1142/S0218126623503218
Ho JLiu M(2024)Research on Document Image Binarization: A Survey2024 IEEE 7th International Conference on Electronic Information and Communication Technology (ICEICT)10.1109/ICEICT61637.2024.10671186(457-462)Online publication date: 31-Jul-2024
https://doi.org/10.1109/ICEICT61637.2024.10671186
Khan SUllah IKhan FLee YUllah S(2023)Historical Text Image Enhancement Using Image Scaling and Generative Adversarial NetworksSensors10.3390/s2308400323:8(4003)Online publication date: 14-Apr-2023
https://doi.org/10.3390/s23084003
Biswas RSarkhel SRoy SPal U(2023)TransDocUNet: A Transformer-based UNet Architecture for Degraded Document Image BinarizationProceedings of the Fourteenth Indian Conference on Computer Vision, Graphics and Image Processing10.1145/3627631.3627639(1-9)Online publication date: 15-Dec-2023
https://dl.acm.org/doi/10.1145/3627631.3627639
Rezanezhad VBaierer KNeudecker C(2023)A hybrid CNN-Transformer model for Historical Document Image BinarizationProceedings of the 7th International Workshop on Historical Document Imaging and Processing10.1145/3604951.3605508(79-84)Online publication date: 25-Aug-2023
https://dl.acm.org/doi/10.1145/3604951.3605508
Yang MXu S(2023)A Novel Degraded Document Binarization Model through Vision Transformer NetworkInformation Fusion10.1016/j.inffus.2022.12.01193(159-173)Online publication date: May-2023
https://doi.org/10.1016/j.inffus.2022.12.011
Ghosh MMukherjee HObaidullah SGao XRoy K(2023)Scene text understanding: recapitulating the past decadeArtificial Intelligence Review10.1007/s10462-023-10530-356:12(15301-15373)Online publication date: 18-Jun-2023
https://doi.org/10.1007/s10462-023-10530-3
Bhowmik SBhowmik S(2023)Document Image BinarizationDocument Layout Analysis10.1007/978-981-99-4277-0_2(11-30)Online publication date: 1-Aug-2023
https://doi.org/10.1007/978-981-99-4277-0_2
Rosello ACastellanos FMartinez-Esteso JGallego ACalvo-Zaragoza J(2023)Test-Time Augmentation for Document Image BinarizationPattern Recognition and Image Analysis10.1007/978-3-031-36616-1_13(158-169)Online publication date: 25-Jun-2023
https://doi.org/10.1007/978-3-031-36616-1_13
Show More Cited By

View Options

Login options

Check if you have access through your login credentials or your institution to get full access on this article.

Full Access

Get this Publication

View options

PDF

View or Download as a PDF file.

eReader

View online with eReader.

Figures

Tables

Media

View Table of Conten