Paper Info
Title
Handwritten Arabic text recognition using multi-stage sub-core-shape HMMs.
Abstract
In this paper, we present a multi-stage HMM-based text recognition system for handwritten Arabic. This system employs a novel way of representing Arabic characters by separating the core shapes from the diacritics and then representing these core shapes by smaller units which we term as sub-core shapes. This results in huge reductions in the number of models that need to be trained for the text recognition task. Further, contextual HMM modeling utilizing these sub-core shapes is presented which demonstrates that using sub-core shapes as models improves the contextual HMM system in comparison with a contextual HMM system employing the standard Arabic character shapes as models, and it leads to significantly compact recognizer at the same time. Furthermore, multi-stream contextual sub-core-shape HMMs are presented where the features computed from a sliding window form one stream and its horizontal derivative features are the second stream with each stream having different weights. The system is evaluated on two publicly available databases for different text recognition tasks including conditions where little training data are available. The presented system outperforms the standard character-shape system on all the text recognition tasks on both the databases.
Year
DOI
Venue
2019
10.1007/s10032-019-00339-8
International Journal on Document Analysis and Recognition (IJDAR)
Keywords
Field
DocType
Handwritten text recognition, Arabic text recognition, Arabic sub-core shapes, Separating core shapes and diacritics, Multi-stage text recognition, Hidden Markov models
Training set,Sliding window protocol,Pattern recognition,Arabic,Computer science,Arabic characters,Artificial intelligence,Hidden Markov model,Text recognition
Journal
Volume
Issue
ISSN
22
3
1433-2833
Citations 
PageRank 
References 
1
0.37
0
Authors
2
Name
Order
Citations
PageRank
Irfan Ahmad112310.13
G. A. Fink211814.89