Yorùbá Character Recognition System Using Convolutional Recurrent Neural Network

Abstract

Handwritten recognition systems enable automatic recognition of human handwritings, thereby increasing human-computer interaction. Despite enormous efforts in handwritten recognition, little progress has been made due to the variability of human handwriting, which presents numerous difficulties for machines to recognize. It was discovered that while tremendous progress has been made in handwritten recognition of English and Arabic languages, very little work has been done on Yorùbá handwritten characters. Those few works, in turn, made use of Hidden Markov Model (HMM), Support Vector Machine (SVM), Bayes theorem, and decision tree algorithms. To integrate and save one of Nigeria's indigenous languages from extinction, as well as to make Yorùbá documents accessible and available in the digital world, this research work was undertaken. The research presents a convolutional recurrent neural network (CRNN) for the recognition of Yorùbá handwritten characters. Data were collected from students of Kwara State University who were literate Yorùbá writers. The collected data were subjected to some level of preprocessing such as grayscale, binarization, and normalization in order to remove perturbations introduced during the digitization process. The convolutional recurrent neural network model was trained using the preprocessed images. The evaluation was conducted using the acquired Yorùbá characters, 87.5% of the acquired images were used for the training while 12.5% were used to evaluate the developed system. As there is currently no publicly available database of Yorùbá characters for validating Yorùbá recognition systems. The resulting recognition accuracy was 87.2% while the characters with under dot and diacritic signs has low recognition accuracy.

Keywords

• Handwriting recognition
• Yorùbá
• Convolutional recurrent neural network
• Document analysis
• Deep learning

References

1. Ajao JF, Olawuyi DO, Odejobi OO. 2018. Yoruba handwritten character recognition using freeman chain code and k-nearest neighbor classifier. J Teknol dan Sist Komp, 6(4): 129-134.
2. Altwaijry N, Al-Turaiki I. 2021. Arabic handwriting recognition system using convolutional neural network. Neural Comp App, 33(7): 2249-2261.
3. Asahiah FO. 2014. Development of a Standard Yorùbá digital text automatic diacritic restoration system. Signature, 22: 02.
4. Bamgbose A. 2000. A grammar of Yorùbá (Vol. 5). Cambridge University Press, Cambridge, UK, pp: 70.
5. Bluche T, Ney H, Kermorvant C. 2014. A comparison of sequence-trained deep neural networks and recurrent neural networks optical modeling for handwriting recognition. In International conference on statistical language and speech processing. Springer, Cham, Germany, pp: 199-210.
6. Calvo-Zaragoza J, Toselli AH, Vidal E. 2019. Handwritten music recognition for mensural notation with convolutional recurrent neural networks. Pattern Recog Lett, 128: 115-121.
7. Chacko BP, Krishnan VV, Raju G, Anto PB. 2012. Handwritten character recognition using wavelet energy and extreme learning machine. Int J Machine Learn Cybernet, 3(2): 149-161.
8. Chaudhari K, Thakkar A. 2019. Survey on handwriting-based personality trait identification. Expert Sys App, 124: 282-308.

9. Darwish K, Habash N, Abbas M, Al-Khalifa H, Al-Natsheh HT, Bouamor H. Mubarak H. 2021. A panoramic survey of natural language processing in the Arab world. Commun ACM, 64(4): 72-81.
10. Das K, Behera RN. 2017. A survey on machine learning: concept, algorithms and applications. Int J Innov Res Comp Commun Eng, 5(2): 1301-1309.
11. Dewa CK, Fadhilah AL. Afiahayati A. 2018. Convolutional neural networks for handwritten Javanese character recognition. Indonesian J Comp Cybernetics Sys, 12(1): 83-94.
12. Garoot AH, Safar M, Suen CY. 2017. A comprehensive survey on handwriting and computerized graphology. In 2017 14th IAPR International Conference on Document Analysis and Recognition (ICDAR), November 9-15, 2017, Kyoto, Japan, Vol. 1, pp: 621-626.
13. Gatos B, Pratikakis I, Perantonis SJ. 2004. An adaptive binarization technique for low quality historical documents. In International Workshop on Document Analysis Systems. Springer, Berlin, Germany, pp: 102-113.
14. Graves A, Liwicki M, Fernández S, Bertolami R, Bunke H, Schmidhuber J. 2008. A novel connectionist system for unconstrained handwriting recognition. IEEE Trans Pattern Anal Machine Intell, 31(5): 855-868.
15. Likforman-Sulem L, Zahour A, Taconet B. 2007. Text line segmentation of historical documents: a survey. International J Doc Anal Recog, 9(2-4): 123-138.
16. Ly NT, Nguyen CT, Nguyen KC, Nakagawa M. 2017. Deep convolutional recurrent network for segmentation-free offline handwritten Japanese text recognition. In 2017 14th IAPR International Conference on Document Analysis and Recognition, November 9-15, 2017, Kyoto, Japan, Vol. 7, pp: 5-9.
17. Mithe R, Indalkar S, Divekar N. 2013. Optical character recognition. Int J Recent Technol Eng, 2(1): 72-75.
18. Nina O, Morse B, Barrett W. 2011. A recursive Otsu thresholding method for scanned document binarization. In 2011 IEEE Workshop on Applications of Computer Vision (WACV), January 5-7, 2011, Kona, HI, US, pp: 307-314.
19. Ojo O. 2007. The Yorùbá in transition: history, values, and modernity. Africa Today, 54(2): 151-152.
20. Oladele MO, Adepoju TM, Olatoke O, Ojo OA. 2020. Offline Yorùbá handwritten word recognition using geometric feature extraction and support vector machine classifier. Malaysian J Comp, 5(2): 504-514.
21. Óní ỌJ, Asahiah FỌ. 2020. Computational modelling of an optical character recognition system for Yorùbá printed text images. Sci African, 9: e00415.
22. Oyedotun OK, Dimililer K. 2016. Pattern recognition: invariance learning in convolutional auto encoder network. Int J Image Graph Signal Proces, 8(3): 19-27.
23. Peel JDY. 2009. A Heterogeneous volume of Yorùbá history and culture-the Yorùbá in transition: history, values, and modernity. Edited by Toyin Falola and Ann Genova. Carolina Academic Press, Durham, NC, US, pp: 498.
24. Srihari SN, Ball G. 2012. An assessment of Arabic handwriting recognition technology. In Guide to OCR for Arabic Scripts, Springer, London, UK, pp: 3-34.
25. Srihari SN, Yang X, Ball GR. 2007. Offline Chinese handwriting recognition: an assessment of current technology. Front Comp Sci China, 1(2): 137-155.
26. Wang J, Wu R, Zhang S. 2021. Robust recognition of Chinese text from cellphone-acquired low-quality identity card images using convolutional recurrent neural network. Sensors Mater, 33(4): 1187-1198.