Browse > Article
http://dx.doi.org/10.4218/etrij.2017-0248

Enhanced technique for Arabic handwriting recognition using deep belief network and a morphological algorithm for solving ligature segmentation  

Essa, Nada (Department of Information Technology, Faculty of Computers and Information Sciences, Mansoura University)
El-Daydamony, Eman (Department of Information Technology, Faculty of Computers and Information Sciences, Mansoura University)
Mohamed, Ahmed Atwan (Department of Information Technology, Faculty of Computers and Information Sciences, Mansoura University)
Publication Information
ETRI Journal / v.40, no.6, 2018 , pp. 774-787 More about this Journal
Abstract
Arabic handwriting segmentation and recognition is an area of research that has not yet been fully understood. Dealing with Arabic ligature segmentation, where the Arabic characters are connected and unconstrained naturally, is one of the fundamental problems when dealing with the Arabic script. Arabic character-recognition techniques consider ligatures as new classes in addition to the classes of the Arabic characters. This paper introduces an enhanced technique for Arabic handwriting recognition using the deep belief network (DBN) and a new morphological algorithm for ligature segmentation. There are two main stages for the implementation of this technique. The first stage involves an enhanced technique of the Sari segmentation algorithm, where a new ligature segmentation algorithm is developed. The second stage involves the Arabic character recognition using DBNs and support vector machines (SVMs). The two stages are tested on the IFN/ENIT and HACDB databases, and the results obtained proved the effectiveness of the proposed algorithm compared with other existing systems.
Keywords
deep belief networks; deep learning; ligatures; morphology; restricted Boltzmann machine;
Citations & Related Records
연도 인용수 순위
  • Reference
1 Y. Alginahi, A survey on Arabic character segmentation, Int. J. Doc. Anal. Recogn. 16 (2013), no. 2, 105-126.   DOI
2 U. Porwal, Z. Shi, and S. Setlur, Machine learning in handwritten Arabic text recognition, Handbook Statistics 31 (2013), 443-470.
3 M. Parvez and S. Mahmoud, Arabic handwriting recognition using structural and syntactic pattern attributes, Pattern Recogn. 46 (2012), no. 1, 141-154.   DOI
4 A. Lawgali, A survey on Arabic character recognition, Int. J. Signal Proces. Image Process. Pattern Recogn. 8 (2015), no. 2, 401-426.   DOI
5 T. Sari, l. Souici, and M. Sellami, Off-line handwritten Arabic character segmentation algorithm: ACSA, Proc. Int. Workshop. Frontiers Handwriting Recogn. (IWFHR), Ontario, Canada, August 6-8, 2002, pp. 452-457.
6 S. Elaiwat, M. Abu-zanona, and F. AL-Zawaideh, A three stages segmentation model for a higher accurate off-line Arabic handwriting recognition, World Comput. Sci. Inform. Technol. J. 2 (2012), no. 3, 98-104.
7 H. Eraqi and S. Abdelazeem, A new efficient graphemes segmentation technique for offline Arabic handwriting, Proc. Int. Conf. Frontiers Handwriting Recogn. (ICFHR), Bari, Italy, September 18-20, 2012, pp. 95-100.
8 A. Lawgali et al., Automatic segmentation for Arabic characters in handwriting documents, Proc. Int. Conf. Image Process., Brussels, Belgium, September 11-14, 2011, pp. 3560-3563.
9 Z. Tamen, H. Drias, and D. Boughaci, An efficient multiple classifier system for Arabic handwritten words recognition, Pattern Recogn. Lett. 39 (2017), 123-132.
10 A. AbdelRaouf et al., Arabic character recognition using a Haar cascade classifier approach (Hcc), Pattern Anal. Appl. 19 (2016), no. 2, 411-426.   DOI
11 M. Elleuch and M. Kherallah, An improved Arabic handwritten recognition system using deep support vector machines, Int. J. Multimed. Data Eng. Manag. 7 (2016), no. 2, 1-20.   DOI
12 U. Porwal, Y. Zhou, and V. Govindaraju, Handwritten Arabic text recognition using deep belief networks, Proc. Int. Conf. Pattern Recogn. (ICPR), Tsukuba, Japan, November 11-15, 2012, pp. 302-305.
13 L. Chergui and M. Kef, SIFT descriptors for Arabic handwriting recognition, Int. J. Comput. Vis. Robot. 5 (2015), no. 4, 441-461.   DOI
14 I. Ahmad and G. Fink, Multi-stage hmm based Arabic text recognition with rescoring, Proc. Int. Conf. Doc. Anal. Recogn. (ICDAR), Nancy, France, August 23-26, 2015, pp. 751-755.
15 M. Kadhm and A. Abdul Hassan, Handwriting word recognition based on SVM classifier, Int. J. Adv. Comput. Sci. Appl. 6 (2015), no. 11, 64-68.
16 K. Jayech, M. Mahjoub, and N. Ben Amara, Arabic handwriting recognition based on synchronous multi-stream hmm without explicit segmentation, Proc. Int. Conf. Hybrid Artif. Intell. Syst., Bilbao, Spain, June 22-24, 2015, pp. 136-145.
17 A. AL-Shatnawi and K. Omar, Methods of Arabic language baseline detection - the state of art, Int. J. Comput. Sci. Netw. Secur. 8 (2008), no. 10, 137-143.
18 N. Habash, Introduction to Arabic natural language processing, Morgan & Claypool, New York, NY, 2010.
19 L. Lam, S. Lee, and C. Suen, Thinning methodologies-a comprehensive survey, IEEE Trans. Pattern Anal. Mach. Intell. 14 (1992), no. 9, 869-885.   DOI
20 A. Zeki, The segmentation problem in Arabic character recognition the state of the art, Proc. Int. Conf. Infom. Commun. Technol. (ICICT), Karachi, Pakistan, August 27-28, 2005, pp. 11-26.
21 T. Abu-Ain et al., Text normalization framework for handwritten cursive languages by detection and straightness the writing baseline, Procedia Technol. 11 (2013), 666-671.   DOI
22 S. Touj, N. Amara, and H. Amiri, Arabic handwritten words recognition based on a planar hidden Markov model, Int. Arab J. Inform. Technol. 2 (2005), no. 4, 318-325.
23 G. Hinton, S. Osindero, and Y. Teh, A Fast learning algorithm for deep belief nets, Neural Comput. 18 (2006), no. 7, 1527-1554.   DOI
24 P. Gader, A. Gillies, and D. Hepp, Handwritten character recognition, in Digital image processing methods (Ed. E. Dougherty), Marcel Dekker, New York, NY, 1994, pp. 246-247.
25 T. Sari and M. Sellami, Overview of some algorithms of off-line Arabic handwriting segmentation, Int. Arab J. Inform. Technol. 4 (2007), no. 4, 289-300.
26 Y. Elarian et al., Arabic ligatures: analysis and application in text recognition, Proc. Int. Conf. Doc. Anal. Recogn., Tunis, Tunisia, August 23-26, 2015, pp. 896-900.
27 N. Dalal and B. Triggs, Histograms of oriented gradients for human detection, Conf. Comput. Vis. Pattern Recogn. (CVPR), San Diego, CA, June 20-25, 2005, pp. 1-8.
28 H. Bay, T. Tuytelaars, and L. Gool, Speeded up robust features (SURF), Proc. Eur. Conf. Comput. Vis., Graz, Austria, May 7-13, 2006, pp. 404-417.
29 M. Carreira-Perpinan, G. Hinton, On contrastive divergence learning, Proc. Int. Workshop. Artif. Intell. Statistics, Barbados, January 6-8, 2005, pp. 33-44.
30 M. Keyvanrad and M. Homayounpour, A brief survey on deep belief networks and introducing a new object oriented toolbox (DeeBNet V3. 0), arXiv: 1408.3264, 2014.
31 C. Cortes and V. Vapnik, Support-vector networks, Mach. Learn 3 (1995), no. 3, 273-297.
32 M. Pechwitz et al., IFN/ENIT-database of handwritten Arabic words, Colloq. Int. Francophone Sur L'ecrit Et Le Doc. (CIFED), Tunisia, October 2002, pp. 127-136.
33 G. Abanda and F. Jamour, A word matching algorithm in handwritten Arabic recognition using multiple-sequence weighted edit distances, Int. J. Comput. Sci. Issues 11 (2014), no. 3, 18-26.
34 A. Lawgali, M. Angelova, and A. Bouridane, HACDB: handwritten Arabic characters database for automatic character recognition, Eur. Workshop. Vis. Inform. Process., Paris, France, June 10-12, 2013, pp. 255-259.
35 J. Davis and M. Goadrich, The relationship between precision-recall and roc curves, Int. Conf. Mach. Learn., Pittsburgh, PA, June 25-29, 2006, 233-240.