Alignment of Historical Handwritten Manuscripts Using Siamese Neural Network

Majeed Kassis; Jumana Nassour; Jihad El-Sana

doi:https://doi.org/10.1109/ICDAR.2017.56

Alignment of Historical Handwritten Manuscripts Using Siamese Neural Network

Majeed Kassis, Jumana Nassour, Jihad El-Sana

Department of Computer Science

Ben-Gurion University of the Negev

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution › peer-review

Abstract

Historical manuscript alignment is a widely known problem in historical document analysis, and the attempt of finding the differences between manuscript editions is mainly done by hand. Today, most of the computational tools coming to assist the historians are based on word recognition or spotting. These solutions are partial at best. In this paper, we present a Siamese neural network based system, which automatically identifies whether a pair of images contain the same text without the need of recognizing the text. The user is required to annotate several pages of two manuscripts, and with the assistance of synthetically generated data and affine distortions we can align two manuscripts written by different writers, achieving strong results.

Original language	American English
Title of host publication	Proceedings - 14th IAPR International Conference on Document Analysis and Recognition, ICDAR 2017
Pages	293-298
Number of pages	6
ISBN (Electronic)	9781538635865
DOIs	https://doi.org/10.1109/ICDAR.2017.56
State	Published - 2 Jul 2017
Event	14th IAPR International Conference on Document Analysis and Recognition, ICDAR 2017 - Kyoto, Japan Duration: 9 Nov 2017 → 15 Nov 2017

Publication series

Name	Proceedings of the International Conference on Document Analysis and Recognition, ICDAR
Volume	1

Conference

Conference	14th IAPR International Conference on Document Analysis and Recognition, ICDAR 2017
Country/Territory	Japan
City	Kyoto
Period	9/11/17 → 15/11/17

Keywords

Deep learning
Manuscript alignment
Siamese network

All Science Journal Classification (ASJC) codes

Computer Vision and Pattern Recognition

Access to Document

https://doi.org/10.1109/ICDAR.2017.56

Cite this

Kassis, M., Nassour, J., & El-Sana, J. (2017). Alignment of Historical Handwritten Manuscripts Using Siamese Neural Network. In Proceedings - 14th IAPR International Conference on Document Analysis and Recognition, ICDAR 2017 (pp. 293-298). (Proceedings of the International Conference on Document Analysis and Recognition, ICDAR; Vol. 1). https://doi.org/10.1109/ICDAR.2017.56

@inproceedings{95560a7d0a084bb3a2149ce1c8ef6dc8,

title = "Alignment of Historical Handwritten Manuscripts Using Siamese Neural Network",

abstract = "Historical manuscript alignment is a widely known problem in historical document analysis, and the attempt of finding the differences between manuscript editions is mainly done by hand. Today, most of the computational tools coming to assist the historians are based on word recognition or spotting. These solutions are partial at best. In this paper, we present a Siamese neural network based system, which automatically identifies whether a pair of images contain the same text without the need of recognizing the text. The user is required to annotate several pages of two manuscripts, and with the assistance of synthetically generated data and affine distortions we can align two manuscripts written by different writers, achieving strong results.",

keywords = "Deep learning, Manuscript alignment, Siamese network",

author = "Majeed Kassis and Jumana Nassour and Jihad El-Sana",

note = "Publisher Copyright: {\textcopyright} 2017 IEEE.; 14th IAPR International Conference on Document Analysis and Recognition, ICDAR 2017 ; Conference date: 09-11-2017 Through 15-11-2017",

year = "2017",

month = jul,

day = "2",

doi = "https://doi.org/10.1109/ICDAR.2017.56",

language = "American English",

series = "Proceedings of the International Conference on Document Analysis and Recognition, ICDAR",

pages = "293--298",

booktitle = "Proceedings - 14th IAPR International Conference on Document Analysis and Recognition, ICDAR 2017",

}

Kassis, M, Nassour, J & El-Sana, J 2017, Alignment of Historical Handwritten Manuscripts Using Siamese Neural Network. in Proceedings - 14th IAPR International Conference on Document Analysis and Recognition, ICDAR 2017. Proceedings of the International Conference on Document Analysis and Recognition, ICDAR, vol. 1, pp. 293-298, 14th IAPR International Conference on Document Analysis and Recognition, ICDAR 2017, Kyoto, Japan, 9/11/17. https://doi.org/10.1109/ICDAR.2017.56

Alignment of Historical Handwritten Manuscripts Using Siamese Neural Network. / Kassis, Majeed; Nassour, Jumana; El-Sana, Jihad.
Proceedings - 14th IAPR International Conference on Document Analysis and Recognition, ICDAR 2017. 2017. p. 293-298 (Proceedings of the International Conference on Document Analysis and Recognition, ICDAR; Vol. 1).

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution › peer-review

TY - GEN

T1 - Alignment of Historical Handwritten Manuscripts Using Siamese Neural Network

AU - Kassis, Majeed

AU - Nassour, Jumana

AU - El-Sana, Jihad

PY - 2017/7/2

Y1 - 2017/7/2

N2 - Historical manuscript alignment is a widely known problem in historical document analysis, and the attempt of finding the differences between manuscript editions is mainly done by hand. Today, most of the computational tools coming to assist the historians are based on word recognition or spotting. These solutions are partial at best. In this paper, we present a Siamese neural network based system, which automatically identifies whether a pair of images contain the same text without the need of recognizing the text. The user is required to annotate several pages of two manuscripts, and with the assistance of synthetically generated data and affine distortions we can align two manuscripts written by different writers, achieving strong results.

AB - Historical manuscript alignment is a widely known problem in historical document analysis, and the attempt of finding the differences between manuscript editions is mainly done by hand. Today, most of the computational tools coming to assist the historians are based on word recognition or spotting. These solutions are partial at best. In this paper, we present a Siamese neural network based system, which automatically identifies whether a pair of images contain the same text without the need of recognizing the text. The user is required to annotate several pages of two manuscripts, and with the assistance of synthetically generated data and affine distortions we can align two manuscripts written by different writers, achieving strong results.

KW - Deep learning

KW - Manuscript alignment

KW - Siamese network

UR - http://www.scopus.com/inward/record.url?scp=85045218023&partnerID=8YFLogxK

U2 - https://doi.org/10.1109/ICDAR.2017.56

DO - https://doi.org/10.1109/ICDAR.2017.56

M3 - Conference contribution

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

SP - 293

EP - 298

BT - Proceedings - 14th IAPR International Conference on Document Analysis and Recognition, ICDAR 2017

T2 - 14th IAPR International Conference on Document Analysis and Recognition, ICDAR 2017

Y2 - 9 November 2017 through 15 November 2017

ER -

Alignment of Historical Handwritten Manuscripts Using Siamese Neural Network

Abstract

Publication series

Conference

Keywords

All Science Journal Classification (ASJC) codes

Access to Document

Other files and links

Fingerprint

Cite this