An End to End System for Online Handwritten Bangla Character Recognition
Date
2022-05-30
Journal Title
Journal ISSN
Volume Title
Publisher
Department of Computer Science and Engineering(CSE), Islamic University of Technology(IUT), Board Bazar, Gazipur, Bangladesh
Abstract
This report summarizes the attempt to find the way towards building an Optical
Character Recognition System for handwritten Bangla characters. The complex
and unique structure of scripts like Bangla and ever challenging nature of hand-
written texts combined makes it really difficult to complete a perfect system to
approach to convert the scanned handwritten Bangla scripts to machine editable
digital counterpart format of it- as segmentation of the whole image into char-
acters and then classification of the segmented characters is difficult enough to
make the task challenging. In our work, we propose to approach the segmentation
process (directly segment to words) with Distance Transform and morphological
operations for error correction later. Then two zone approach (either side of
matra- upper and lower zone) and apply connected component analysis on both
zones. We handled or adjusted the failed and not directly successful cases by
experimenting with the characteristics of handwritten characters. Then for clas-
sification process, we proposed to classify the segmented characters using neural
networks trained on the relatively newly available datasets. Multiple column,
Mixed characters (Bangla- other languages) and Scene Text Recognition is out
of the scope of our study so far. And we could not include the post-processing
part for our work for lack of work or mention in existing literature, which might
be a great addition in the way of building a complete OCR system.
Description
Supervised by
Mr. A.B.M. Ashiqur Rahman,
Assistant Professor,
Co-Supervisor:
Shahriar Ivan
Lecturer,
Department of Computer Science and Engineering(CSE),
Islamic University of Technology (IUT)
Board Bazar, Gazipur-1704, Bangladesh.
This thesis is submitted in partial fulfillment of the requirements for the degree of Bachelor of Science in Computer Science and Engineering, 2022.
Keywords
Bangla Handwritten Character Recognition, Handwritten Doc- ument Recognition, Optical character recognition
Citation
[1] B. B. Chaudhuri and U. Pal, “A complete printed Bangla OCR system,” Pattern Recognit., vol. 31, no. 5, pp. 531–549, 1998, doi: 10.1016/S0031-3203(97)00078-2. [2] F. Yeasmin Omee, S. Shabbir Himel, and A. Naser Bikas, “A Complete Workflow for Development of Bangla OCR,” Int. J. Comput. Appl., vol. 21, no. 9, pp. 1–6, 2011, doi: 10.5120/2543-3483. [3] A. K. M. S. A. Rabby, S. Haque, M. S. Islam, S. Abujar, and S. A. Hossain, Ekush: A Multipurpose and Multitype Comprehensive Database for Online Off-Line Bangla Handwritten Characters, vol. 1037. Springer Singapore, 2019. [4] J. Ferdous, S. Karmaker, A. K. M. S. A. Rabby, and S. A. Hossain, “MatriVasha: A Multipurpose Comprehensive Database for Bangla Handwritten Compound Characters,” Lect. Notes Networks Syst., vol. 164, pp. 813–821, 2021, doi: 10.1007/978-981-15-9774-9_74 [5] M. F. Mridha, A. Q. Ohi, M. A. Ali, M. I. Emon, and M. M. Kabir, “BanglaWriting : A multi-purpose offline Bangla handwriting dataset,” Data Br., vol. 34, p. 106633, 2021, doi: 10.1016/j.dib.2020.106633. [6] P. P. Roy, A. K. Bhunia, A. Das, P. Dey, and U. Pal, “HMM-based Indic handwritten word recognition using zone segmentation,” Pattern Recognit., vol. 60, pp. 1057– 1075, 2016, doi: 10.1016/j.patcog.2016.04.012. [7] S. Malakar, R. Sarkar, S. Basu, M. Kundu, and M. Nasipuri, “An image database of handwritten Bangla words with automatic benchmarking facilities for character segmentation algorithms,” Neural Comput. Appl., vol. 33, no. 1, pp. 449–468, 2021, doi: 10.1007/s00521-020-04981-w. [8] S. Basu, R. Sarkar, N. Das, M. Kundu, M. Nasipuri, and D. K. Basu, “A fuzzy technique for segmentation of handwritten Bangla word images,” Proc. - Int. Conf. Comput. Theory Appl. ICCTA 2007, no. March, pp. 427–432, 2007, doi: 10.1109/ICCTA.2007.7. 38 [9] R. Sarkar, N. Das, S. Basu, M. Kundu, M. Nasipuri, and D. K. Basu, “CMATERdb1: A database of unconstrained handwritten Bangla and Bangla-English mixed script document image,” Int. J. Doc. Anal. Recognit., vol. 15, no. 1, pp. 71–83, 2012, doi: 10.1007/s10032-011-0148-6. [10] Y. Park, “Discrete Hough transform using line segment representation for line detection,” Opt. Eng., vol. 50, no. 8, p. 087004, 2011, doi: 10.1117/1.3607414. [11] V. Papavassiliou, T. Stafylakis, V. Katsouros, and G. Carayannis, “Handwritten document image segmentation into text lines and words,” Pattern Recognit., vol. 43, no. 1, pp. 369–377, 2010, doi: 10.1016/j.patcog.2009.05.007. [12] A. K. M. Shahariar, A. Rabby, S. Haque, S. Abujar, and S. A. Hossain, “EkushNet : Using Convolutional Neural Network for Bangla Handwritten character recognition” Procedia Comput. Sci., vol. 143, no. December, pp. 603–610, 2018, doi: 10.1016/j.procs.2018.10.437. [13] S. Haque, S. Abujar, S. Abujar, S. Akhter, and S. A. Hossain, “BornoNet : Bangla Handwritten Characters Recognition Using Convolutional Neural Network Convolutional Neural Network,” vol. 00, 2018. [14] S. Irfan and A. Meerza, “Performance Evaluation of Different Algorithms for Handwritten Isolated Bangla Character Recognition,” 2019 Int. Conf. Robot. Signal Process. Tech., pp. 412–416, 2019. [15] M. Biswas et al., “BanglaLekha-Isolated: A Comprehensive Bangla Handwritten Character Dataset,” pp. 1–4, 2017, [Online]. Available: http://arxiv.org/abs/1703. 10661. [16] R. Sarkar, S. Malakar, N. Das, S. Basu, M. Kundu, and M. Nasipuri, “Word extraction and character segmentation from text lines of unconstrained handwritten Bangla document images,” J. Intell. Syst., vol. 20, no. 3, pp. 227–260, 2011, doi: 10.1515/JISYS.2011.013. 39 [17] P. K. Singh, S. Sinha, S. P. Chowdhury, R. Sarkar, and M. Nasipuri, “Word segmentation from unconstrained handwritten Bangla document images using distance transform,” 6th Int. Conf. Adv. Comput. Control. Telecommun. Technol. ACT 2015, pp. 473–484, 2015, doi: 10.1515/9783110450101-041. [18] U. Pal and S. Datta, “Segmentation of Bangla unconstrained handwritten text,” Proc. Int. Conf. Doc. Anal. Recognition, ICDAR, vol. 2003-Janua, no. August, pp. 1128–1132, 2003, doi: 10.1109/ICDAR.2003.1227832. [19] M. Z. Alom, P. Sidike, M. Hasan, T. M. Taha, and V. K. Asari, “Handwritten Bangla Character Recognition Using the State-of-the-Art Deep Convolutional Neural Networks,” Comput. Intell. Neurosci., vol. 2018, pp. 1–12, 2018, doi: 10.1155/2018/6747098. [20] D. Liu and J. Yu, “Otsu method and K-means,” Proc. - 2009 9th Int. Conf. Hybrid Intell. Syst. HIS 2009, vol. 1, no. 2, pp. 344–349, 2009, doi: 10.1109/HIS.2009.74. [21] H. Fan, L. Zhu, and Y. Tang, “Skew detection in document images based on rectangular active contour,” Int. J. Doc. Anal. Recognit., vol. 13, no. 4, pp. 261–269, 2010, doi: 10.1007/s10032-010-0119-3. [22] N. Ouwayed and A. Belaত্তd, “A general approach for multi-oriented text line extraction of handwritten documents,” Int. J. Doc. Anal. Recognit., vol. 15, no. 4, pp. 297–314, 2012, doi: 10.1007/s10032-011-0172-6. [23] L. Zhou, Y. Lu, and C. L. Tan, “Bangla/English script identification based on analysis of connected component profiles,” Lect. Notes Comput. Sci. (including Subser. Lect. Notes Artif. Intell. Lect. Notes Bioinformatics), vol. 3872 LNCS, pp. 243–254, 2006, doi: 10.1007/11669487_22. [24] “GitHub - subashis-dev/Bangla-handwritten-word-segmentation-from-document: Segment all the words from a Bengali handwritten document easily.” https://gith ub.com/subashis-dev/Bangla-handwritten-word-segmentation-from-document (accessed Feb. 23, 2022). 40 [25] H. Kong, “A MEDIAL AXIS BASED THINNING STRATEGY AND STRUCTURAL FEATURE Soumen Bag and Gaurav Harit Department of Computer Science and Engineering Indian Institute of Technology Kharagpur , Kharagpur-721 302 , India,” pp. 2173–2176, 2010. [26] B. B. Chaudhuri, “Corpus-based empirical analysis of form , function and frequency of characters used in Bangla.” [27] “Letter frequency.” http://simia.net/letters/ (accessed Mar. 03, 2022).