Deep BLSTM neural networks for unconstrained continuous handwritten text recognition

Volkmar Frinken, Seiichi Uchida

Research output: Chapter in Book/Report/Conference proceedingConference contribution

12 Citations (Scopus)

Abstract

Recently, two different trends in neural network-based machine learning could be observed. The first one are the introduction of Bidirectional Long Short-Term Memory (BLSTM) neural networks (NN) which made sequences with long-distant dependencies amenable for neural network-based processing. The second one are deep learning techniques, which greatly increased the performance of neural networks, by making use of many hidden layers. In this paper, we propose to combine these two ideas for the task of unconstrained handwriting recognition. Extensive experimental evaluation on the IAM database demonstrate an increase of the recognition performance when using deep learning approaches over commonly used BLSTM neural networks, as well as insight into how different types of hidden layers affect the recognition accuracy.

Original languageEnglish
Title of host publication13th IAPR International Conference on Document Analysis and Recognition, ICDAR 2015 - Conference Proceedings
PublisherIEEE Computer Society
Pages911-915
Number of pages5
ISBN (Electronic)9781479918058
DOIs
Publication statusPublished - Nov 20 2015
Event13th International Conference on Document Analysis and Recognition, ICDAR 2015 - Nancy, France
Duration: Aug 23 2015Aug 26 2015

Publication series

NameProceedings of the International Conference on Document Analysis and Recognition, ICDAR
Volume2015-November
ISSN (Print)1520-5363

Other

Other13th International Conference on Document Analysis and Recognition, ICDAR 2015
CountryFrance
CityNancy
Period8/23/158/26/15

Fingerprint

Neural networks
Learning systems
Long short-term memory
Processing
Deep learning

All Science Journal Classification (ASJC) codes

  • Computer Vision and Pattern Recognition

Cite this

Frinken, V., & Uchida, S. (2015). Deep BLSTM neural networks for unconstrained continuous handwritten text recognition. In 13th IAPR International Conference on Document Analysis and Recognition, ICDAR 2015 - Conference Proceedings (pp. 911-915). [7333894] (Proceedings of the International Conference on Document Analysis and Recognition, ICDAR; Vol. 2015-November). IEEE Computer Society. https://doi.org/10.1109/ICDAR.2015.7333894

Deep BLSTM neural networks for unconstrained continuous handwritten text recognition. / Frinken, Volkmar; Uchida, Seiichi.

13th IAPR International Conference on Document Analysis and Recognition, ICDAR 2015 - Conference Proceedings. IEEE Computer Society, 2015. p. 911-915 7333894 (Proceedings of the International Conference on Document Analysis and Recognition, ICDAR; Vol. 2015-November).

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Frinken, V & Uchida, S 2015, Deep BLSTM neural networks for unconstrained continuous handwritten text recognition. in 13th IAPR International Conference on Document Analysis and Recognition, ICDAR 2015 - Conference Proceedings., 7333894, Proceedings of the International Conference on Document Analysis and Recognition, ICDAR, vol. 2015-November, IEEE Computer Society, pp. 911-915, 13th International Conference on Document Analysis and Recognition, ICDAR 2015, Nancy, France, 8/23/15. https://doi.org/10.1109/ICDAR.2015.7333894
Frinken V, Uchida S. Deep BLSTM neural networks for unconstrained continuous handwritten text recognition. In 13th IAPR International Conference on Document Analysis and Recognition, ICDAR 2015 - Conference Proceedings. IEEE Computer Society. 2015. p. 911-915. 7333894. (Proceedings of the International Conference on Document Analysis and Recognition, ICDAR). https://doi.org/10.1109/ICDAR.2015.7333894
Frinken, Volkmar ; Uchida, Seiichi. / Deep BLSTM neural networks for unconstrained continuous handwritten text recognition. 13th IAPR International Conference on Document Analysis and Recognition, ICDAR 2015 - Conference Proceedings. IEEE Computer Society, 2015. pp. 911-915 (Proceedings of the International Conference on Document Analysis and Recognition, ICDAR).
@inproceedings{3bdac393c1e643908b4f9ba9ad88fd02,
title = "Deep BLSTM neural networks for unconstrained continuous handwritten text recognition",
abstract = "Recently, two different trends in neural network-based machine learning could be observed. The first one are the introduction of Bidirectional Long Short-Term Memory (BLSTM) neural networks (NN) which made sequences with long-distant dependencies amenable for neural network-based processing. The second one are deep learning techniques, which greatly increased the performance of neural networks, by making use of many hidden layers. In this paper, we propose to combine these two ideas for the task of unconstrained handwriting recognition. Extensive experimental evaluation on the IAM database demonstrate an increase of the recognition performance when using deep learning approaches over commonly used BLSTM neural networks, as well as insight into how different types of hidden layers affect the recognition accuracy.",
author = "Volkmar Frinken and Seiichi Uchida",
year = "2015",
month = "11",
day = "20",
doi = "10.1109/ICDAR.2015.7333894",
language = "English",
series = "Proceedings of the International Conference on Document Analysis and Recognition, ICDAR",
publisher = "IEEE Computer Society",
pages = "911--915",
booktitle = "13th IAPR International Conference on Document Analysis and Recognition, ICDAR 2015 - Conference Proceedings",
address = "United States",

}

TY - GEN

T1 - Deep BLSTM neural networks for unconstrained continuous handwritten text recognition

AU - Frinken, Volkmar

AU - Uchida, Seiichi

PY - 2015/11/20

Y1 - 2015/11/20

N2 - Recently, two different trends in neural network-based machine learning could be observed. The first one are the introduction of Bidirectional Long Short-Term Memory (BLSTM) neural networks (NN) which made sequences with long-distant dependencies amenable for neural network-based processing. The second one are deep learning techniques, which greatly increased the performance of neural networks, by making use of many hidden layers. In this paper, we propose to combine these two ideas for the task of unconstrained handwriting recognition. Extensive experimental evaluation on the IAM database demonstrate an increase of the recognition performance when using deep learning approaches over commonly used BLSTM neural networks, as well as insight into how different types of hidden layers affect the recognition accuracy.

AB - Recently, two different trends in neural network-based machine learning could be observed. The first one are the introduction of Bidirectional Long Short-Term Memory (BLSTM) neural networks (NN) which made sequences with long-distant dependencies amenable for neural network-based processing. The second one are deep learning techniques, which greatly increased the performance of neural networks, by making use of many hidden layers. In this paper, we propose to combine these two ideas for the task of unconstrained handwriting recognition. Extensive experimental evaluation on the IAM database demonstrate an increase of the recognition performance when using deep learning approaches over commonly used BLSTM neural networks, as well as insight into how different types of hidden layers affect the recognition accuracy.

UR - http://www.scopus.com/inward/record.url?scp=84962480779&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84962480779&partnerID=8YFLogxK

U2 - 10.1109/ICDAR.2015.7333894

DO - 10.1109/ICDAR.2015.7333894

M3 - Conference contribution

AN - SCOPUS:84962480779

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

SP - 911

EP - 915

BT - 13th IAPR International Conference on Document Analysis and Recognition, ICDAR 2015 - Conference Proceedings

PB - IEEE Computer Society

ER -