Holistic Persian Handwritten Word Recognition Using Convolutional Neural Network

Document Type : Original Article

Authors

Computer Engineering Department, Kosar University of Bojnord, Bojnord, Iran

Abstract

Due to the cursive-ness and high variability of Persian scripts, the segmentation of handwritten words into sub-words is still a challenging task. These issues could be addressed in a holistic approach by sidestepping segmentation at the character level. In this paper, an end-to-end holistic method based on deep convolutional neural network is proposed to recognize off-line Persian handwritten words. The proposed model uses only five convolutional layers and two fully connected layers for classifying word images effectively, which can lead to a substantial
reduction in parameters. The effect of various pooling strategies is also investigated in this paper. The primary goal of this article is to ignore handcrafted feature extraction and attain a generalized and stable word recognition system. The presented model is assessed using two famous handwritten Persian word databases called Sadri and IRANSHAHR. The recognition accuracies were obtained at 98.6% and 94.6%, on Sadri and IRANSHAHR datasets respectively, and outperformed the state-of-the-art methods.

Keywords

References

1.     Das, D., Nayak, D.R., Dash, R., Majhi, B. and Zhang, Y.-D., "H-wordnet: A holistic convolutional neural network approach for handwritten word recognition", IET Image Processing,  Vol. 14, No. 9, (2020), 1794-1805.DOI: 10.1049/iet-ipr.2019.1398.
2.     Imani, Z., Ahmadyfard, Z. and Zohrevand, A., "Holistic farsi handwritten word recognition using gradient features", Journal of AI and Data Mining,  Vol. 4, No. 1, (2016), 19-25.DOI: 10.5829/idosi.JAIDM.2016.04.01.03.
3.     Akbari, Y., Jalili, M.J., Sadri, J., Nouri, K., Siddiqi, I. and Djeddi, C., "A novel database for automatic processing of persian handwritten bank checks", Pattern Recognition,  Vol. 74, No., (2018), 253-265.DOI: https://doi.org/10.1016/j.patcog.2017.09.011.
4.     Ye, M., Viola, P., Raghupathy, S., Sutanto, H. and Li, C., "Learning to group text lines and regions in freeform handwritten notes", in Ninth International Conference on Document Analysis and Recognition (ICDAR 2007). Vol. 1, No., (2007), 28-32.DOI: 10.1109/ICDAR.2007.4378670.
5.     Razzak, I., Kamran, I. and Naz, S., "Deep analysis of handwritten notes for early diagnosis of neurological disorders", in 2020 International Joint Conference on Neural Networks (IJCNN). Vol., No., (2020), 1-6.DOI: 10.1109/IJCNN48605.2020.9207087.
6.     Vajda, S., Roy, K., Pal, U., Chaudhuri, B.B. and Belaid, A., "Automation of indian postal documents written in bangla and english", International Journal of Pattern Recognition and Artificial Intelligence,  Vol. 23, No. 08, (2009), 1599-1632.DOI: https://doi.org/10.1142/S0218001409007739.
7.     Venu, G. and Xue, H., "Fast handwriting recognition for indexing historical documents", in First International Workshop on Document Image Analysis for Libraries, 2004. Proceedings. Vol., No., (2004), 314-320.DOI: 10.1109/DIAL.2004.1263260.
8.     Sánchez, J.A., Romero, V., Toselli, A.H., Villegas, M. and Vidal, E., "A set of benchmarks for handwritten text recognition on historical documents", Pattern Recognition,  Vol. 94, No., (2019), 122-134.DOI: https://doi.org/10.1016/j.patcog.2019.05.025.
9.     Bhowmik, S., Malakar, S., Sarkar, R., Basu, S., Kundu, M. and Nasipuri, M., "Off-line bangla handwritten word recognition: A holistic approach", Neural Computing and Applications,  Vol. 31, No. 10, (2019), 5783-5798.DOI: 10.1007/s00521-018-3389-1.
10.   Abbaszadeh Arani, S.A.A., Kabir, E. and Ebrahimpour, R., "Combining rtl and ltr hmms to recognise handwritten farsi words of small‐and medium‐sized vocabularies", IET Computer Vision,  Vol. 12, No. 6, (2018), 925-932.DOI: 10.1049/iet-cvi.2017.0645.
11.   Arani, S.A.A.A., Kabir, E. and Ebrahimpour, R., "Handwritten farsi word recognition using nn-based fusion of hmm classifiers with different types of features", International Journal of Image and Graphics,  Vol. 19, No. 01, (2019), 1950001.DOI: https://doi.org/10.1142/S0219467819500013.
12.   Guo, Y., Liu, Y., Oerlemans, A., Lao, S., Wu, S. and Lew, M.S., "Deep learning for visual understanding: A review", Neurocomputing,  Vol. 187, No., (2016), 27-48.DOI: https://doi.org/10.1016/j.neucom.2015.09.116.
13.   Bayesteh, E., Ahmadifard, A. and Khosravi, H., "A lexicon reduction method based on clustering word images in offline farsi handwritten word recognition systems", in 2011 7th Iranian Conference on Machine Vision and Image Processing. Vol., No., (2011), 1-5.DOI: 10.1109/IranianMVIP.2011.6121550.
14.   Sadri, J., Yeganehzad, M.R. and Saghi, J., "A novel comprehensive database for offline persian handwriting recognition", Pattern Recognition,  Vol. 60, No., (2016), 378-393.DOI: https://doi.org/10.1016/j.patcog.2016.03.024.
15.   Dehghan, M., Faez, K., Ahmadi, M. and Shridhar, M., "Handwritten farsi (arabic) word recognition: A holistic approach using discrete hmm", Pattern Recognition,  Vol. 34, No. 5, (2001), 1057-1065.DOI: https://doi.org/10.1016/S0031-3203(00)00051-0.
16.   Dehghan, M., Faez, K., Ahmadi, M. and Shridhar, M., "Unconstrained farsi handwritten word recognition using fuzzy vector quantization and hidden markov models", Pattern Recognition Letters,  Vol. 22, No. 2, (2001), 209-214.DOI: https://doi.org/10.1016/S0167-8655(00)00090-8.
17.   Mozaffari, S., Faez, K., Märgner, V. and El-Abed, H., "Lexicon reduction using dots for off-line farsi/arabic handwritten word recognition", Pattern Recognition Letters,  Vol. 29, No. 6, (2008), 724-734.DOI: https://doi.org/10.1016/j.patrec.2007.11.009.
18.   Broumandnia, A., Shanbehzadeh, J. and Rezakhah Varnoosfaderani, M., "Persian/arabic handwritten word recognition using m-band packet wavelet transform", Image and Vision Computing,  Vol. 26, No. 6, (2008), 829-842.DOI: https://doi.org/10.1016/j.imavis.2007.09.004.
19.   Imani, Z., Ahmadyfard, A. and Zohrevand, A., "Introduction to database farsa: Digital image of handwritten farsi words (in persian)", in 11th Iranian Conference on Intelligent Systems in Persian, Tehran, Iran. Vol., No., (2013).DOI: https://civilica.com/doc/214715/.
20.   Tavoli, R., Keyvanpour, M. and Mozaffari, S., "Statistical geometric components of straight lines (sgcsl) feature extraction method for offline arabic/persian handwritten words recognition", IET Image Processing,  Vol. 12, No. 9, (2018), 1606-1616.
21.   Moghaddam, R.F., Cheriet, M., Adankon, M.M., Filonenko, K. and Wisnovsky, R., "Ibn sina: A database for research on processing and understanding of arabic manuscripts images", in Proceedings of the 9th IAPR International Workshop on Document Analysis Systems, Boston, Massachusetts, USA, Association for Computing Machinery. Vol., No., (2010 of Conference), 11–18.
22.   Pechwitz, M., Maddouri, S.S., Märgner, V., Ellouze, N. and Amiri, H., "Ifn/enit-database of handwritten arabic words", in Proc. of CIFED, Citeseer. Vol. 2, No., (2002), 127-136.
23.   LeCun, Y. and Bengio, Y., "Convolutional networks for images, speech, and time series", The handbook of brain theory and neural networks,  Vol. 3361, No. 10, (1995), 1995.
24.   Liu, L., Ouyang, W., Wang, X., Fieguth, P., Chen, J., Liu, X. and Pietikäinen, M., "Deep learning for generic object detection: A survey", International Journal of Computer Vision,  Vol. 128, No. 2, (2020), 261-318.DOI: 10.1007/s11263-019-01247-4.
25.   Zohrevand, A., Imani, Z. and Ezoji, M., "Deep convolutional neural network for finger-knuckle-print recognition", International Journal of Engineering, Transactions A: Basics,  Vol. 34, No. 7, (2021), 1684-1693.DOI: 10.5829/ije.2021.34.07a.12.
26.   Guo, G. and Zhang, N., "A survey on deep learning based face recognition", Computer Vision and Image Understanding,  Vol. 189, No., (2019), 102805.DOI: https://doi.org/10.1016/j.cviu.2019.102805.
27.   Zohrevand, A., Sattari, M., Sadri, J., Imani, Z., Suen, C.Y. and Djeddi, C., "Comparison of persian handwritten digit recognition in three color modalities using deep neural networks, Cham, Springer International Publishing. Vol., No., (2020), 125-136.DOI: 10.1007/978-3-030-59830-3_11.
28.   Safarzadeh, V.M. and Jafarzadeh, P., "Offline persian handwriting recognition with cnn and rnn-ctc", in 2020 25th International Computer Conference, Computer Society of Iran (CSICC). Vol., No., (2020), 1-10.DOI: 10.1109/CSICC49403.2020.9050073.
29.   Bonyani, M., Jahangard, S. and Daneshmand, M., "Persian handwritten digit, character and word recognition using deep learning", International Journal on Document Analysis and Recognition (IJDAR),  Vol. 24, No. 1, (2021), 133-143.DOI: 10.1007/s10032-021-00368-2.
30.   Huang, G., Liu, Z., Van Der Maaten, L. and Weinberger, K.Q., "Densely connected convolutional networks", in Proceedings of the IEEE conference on computer vision and pattern recognition. Vol., No., (2017), 4700-4708.
31.   Chollet, F., "Xception: Deep learning with depthwise separable convolutions", in Proceedings of the IEEE conference on computer vision and pattern recognition. Vol., No., (2017), 1251-1258.
32.   Sabzi, R., Fotoohinya, Z., Khalili, A., Golzari, S., Salkhorde, Z., Behravesh, S. and Akbarpour, S., "Recognizing persian handwritten words using deep convolutional networks", in 2017 Artificial Intelligence and Signal Processing Conference (AISP). Vol., No., (2017), 85-90.DOI: 10.1109/AISP.2017.8324114.
33.   Scherer, D., Müller, A. and Behnke, S., "Evaluation of pooling operations in convolutional architectures for object recognition", in Artificial Neural Networks – ICANN 2010, Berlin, Heidelberg, Springer Berlin Heidelberg. Vol., No., (2010), 92-101.DOI: https://doi.org/10.1007/978-3-642-15825-4_10.
34.   Zhuang, F., Qi, Z., Duan, K., Xi, D., Zhu, Y., Zhu, H., Xiong, H. and He, Q., "A comprehensive survey on transfer learning", Proceedings of the IEEE,  Vol. 109, No. 1, (2021), 43-76.DOI: 10.1109/JPROC.2020.3004555.
 
International Journal of Engineering
Volume 34, Issue 8
TRANSACTIONS B: Applications
August 2021
Pages 2028-2037

Files

Cited by

Share

How to cite

Statistics