Offline cursive Bengali word recognition using CNNs with a recurrent model

Adak, C; Chaudhuri, BB; Blumenstein, M

Offline cursive Bengali word recognition using CNNs with a recurrent model

Adak, C

Chaudhuri, BB Blumenstein, M

Permalink

Publication Type:: Conference Proceeding
Citation:: Proceedings of International Conference on Frontiers in Handwriting Recognition, ICFHR, 2016, 0 pp. 429 - 434
Issue Date:: 2016-07-02

Closed Access

	Filename	Description	Size
	07814102.pdf	Published version	370.05 kB	Adobe PDF	View/Open

Copyright Clearance Process

Recently Added
In Progress
Closed Access

This item is closed access and not available.

Full metadata record

Field	Value	Language
dc.contributor.author	Adak, C https://orcid.org/0000-0002-9085-2770	en_US
dc.contributor.author	Chaudhuri, BB	en_US
dc.contributor.author	Blumenstein, M https://orcid.org/0000-0002-9908-3744	en_US
dc.date.issued	2016-07-02	en_US
dc.identifier.citation	Proceedings of International Conference on Frontiers in Handwriting Recognition, ICFHR, 2016, 0 pp. 429 - 434	en_US
dc.identifier.isbn	9781509009817	en_US
dc.identifier.issn	2167-6445	en_US
dc.identifier.uri	http://hdl.handle.net/10453/126455
dc.description.abstract	© 2016 IEEE. This paper deals with offline handwritten word recognition of a major Indic script: Bengali. Due to the structure of this script, the characters (mostly ortho-syllables) are frequently overlapping and hard to segment, especially when the writing is cursive. Individual character recognition and the combination of outputs can increase the likelihood of errors. Instead, a better approach can be sending the whole word to a suitable recognizer. Here we use the Convolutional Neural Network (CNN) integrated with a recurrent model for this purpose. Long short-term memory blocks are used as hidden units. Also, the CNN-derived features are employed in a recurrent model with a CTC (Connectionist Temporal Classification) layer to get the output. We have tested our method on three datasets: (a) a publicly available dataset, (b) a new dataset generated by our research group and (c) an unconstrained dataset. The dataset (a) contains 17,091 words, while our dataset (b) contains 107,550 number of words in total. In addition to these, the dataset (c) is comprised of 5,223 words. We have compared our results with those of some earlier work in the area and have found improved performance, which is due to the novel integration of CNNs with the recurrent model.	en_US
dc.relation.ispartof	Proceedings of International Conference on Frontiers in Handwriting Recognition, ICFHR	en_US
dc.relation.isbasedon	10.1109/ICFHR.2016.0086	en_US
dc.title	Offline cursive Bengali word recognition using CNNs with a recurrent model	en_US
dc.type	Conference Proceeding
utslib.citation.volume	0	en_US
utslib.for	0801 Artificial Intelligence and Image Processing	en_US
pubs.embargo.period	Not known	en_US
pubs.organisational-group	/University of Technology Sydney
pubs.organisational-group	/University of Technology Sydney/Faculty of Engineering and Information Technology
pubs.organisational-group	/University of Technology Sydney/Faculty of Engineering and Information Technology/School of Computer Science
pubs.organisational-group	/University of Technology Sydney/Strength - CAI - Centre for Artificial Intelligence
pubs.organisational-group	/University of Technology Sydney/Strength - QSI - Centre for Quantum Software and Information
utslib.copyright.status	closed_access
pubs.publication-status	Published	en_US
pubs.volume	0	en_US

Abstract:

© 2016 IEEE. This paper deals with offline handwritten word recognition of a major Indic script: Bengali. Due to the structure of this script, the characters (mostly ortho-syllables) are frequently overlapping and hard to segment, especially when the writing is cursive. Individual character recognition and the combination of outputs can increase the likelihood of errors. Instead, a better approach can be sending the whole word to a suitable recognizer. Here we use the Convolutional Neural Network (CNN) integrated with a recurrent model for this purpose. Long short-term memory blocks are used as hidden units. Also, the CNN-derived features are employed in a recurrent model with a CTC (Connectionist Temporal Classification) layer to get the output. We have tested our method on three datasets: (a) a publicly available dataset, (b) a new dataset generated by our research group and (c) an unconstrained dataset. The dataset (a) contains 17,091 words, while our dataset (b) contains 107,550 number of words in total. In addition to these, the dataset (c) is comprised of 5,223 words. We have compared our results with those of some earlier work in the area and have found improved performance, which is due to the novel integration of CNNs with the recurrent model.

Please use this identifier to cite or link to this item:

http://hdl.handle.net/10453/126455