Two Stage Multi Modal Deep Learning Kannada Character Recognition Model Adaptive to Discriminative Patterns of Kannada Characters

S R Siddanna; Y C Kiran

doi:10.17485/IJST/v16i3.1904

Article

Two Stage Multi Modal Deep Learning Kannada Character Recognition Model Adaptive to Discriminative Patterns of Kannada Characters

VIEWS 897
PDF 151

Indian Journal of Science and Technology

DOI: 10.17485/IJST/v16i3.1904

Year: 2023, Volume: 16, Issue: 3, Pages: 155-166

Original Article

Two Stage Multi Modal Deep Learning Kannada Character Recognition Model Adaptive to Discriminative Patterns of Kannada Characters

S R Siddanna^1,2*, Y C Kiran²

¹Department of Information Science and Engg, SJB Institute of Technology, Bangalore, 560060, Karnataka, India
2Department of Information Science and Engg, Global Academy of Technology, , Bangalore, 560098, Karnataka, India

*Corresponding Author
Email: [email protected]

Received Date:21 September 2022, Accepted Date:14 December 2022, Published Date:20 January 2023

This work is licensed under a Creative Commons Attribution 4.0 International License.

Abstract

Objectives: Designing optical character recognition systems for Kannada character is challenging due to higher self-similarity in characters and higher number of character classes. This work addresses the two major problems of reduced accuracy and higher false positives due to higher self-similarity in characters. Methods: This work proposes a two stage multi modal deep learning technique to handle the complexity in Kannada character recognition. The characters are first grouped based on morphological and structural similarity. A novel morphological/structural difference maximization convolution kernel based deep learning modal is trained for each of character group to recognize characters in that group. This divide and conquer strategy reduce the complexity of deep learning model in learning discriminative features for Kannada character recognition. Findings: The proposed two stage multi modal deep learning provides 89% recognition accuracy which is at least 6% higher compared to existing works. The false positives in proposed solution are at least10% lower compared to existing works. Novelty: A novel sector intensity distribution feature specific to curve structure of Kannada characters for deciding the number of groups of characters. Classifier is designed for each group of characters. Classification is done using convolutional neural network with a novel morphological/structural difference maximization convolution kernel to solve the structural similarity problem in Kannada character recognition.

Keywords: Handwritten character Recognition; Deep Learning; Machine Learning; Pattern Recognition; Fuzzy Gaussian

References

Hegde P, Rajath S, Shwetha D, Sindhu C, Ravi P. Handwritten Character Recognition of Kannada Language Using Convolutional Neural Networks and Transfer Learning. 2021. Available from: https://doi.org/10.1088/1757-899X/1110/1/012003
Rao AS, S S, K A, Arpitha, Nayak C, Meghana, et al. Exploring Deep Learning Techniques for Kannada Handwritten Character Recognition: A Boon for Digitization. 2020. Available from: http://sersc.org/journals/index.php/IJAST/article/view/25189
Rani NS, Subramani AC, P. AK, Pushpa BR. Deep Learning Network Architecture based Kannada Handwritten Character Recognition. 2020 Second International Conference on Inventive Research in Computing Applications (ICIRCA). 2020. Available from: https://doi.org/10.1109/ICIRCA48905.2020.9183160
Ravikumar M, Sampathkumar S. Recognition of Kannada Handwritten Words from Answer Scripts Using Machine Learning Approaches.Information and Communication Technology for Competitive Strategies (ICTCS 2020) Lecture Notes in Networks and Systems. 2022;191. Available from: https://doi.org/10.1007/978-981-16-0739-4_99
Ramesh G, Sharma GN, Balaji JM, Champa HN. Offline Kannada Handwritten Character Recognition Using Convolutional Neural Networks. 2019 IEEE International WIE Conference on Electrical and Computer Engineering (WIECON-ECE). 2019;p. 1–5. Available from: https://doi.org/10.1109/WIECON-ECE48653.2019.9019914
Gandhana MH, Lakshman D, Naik. Online Kannada Handwritten Characters and Numerical Recognition using CNN Classifier. International Journal Of Engineering Research & Technology . 2021;10(07). Available from: https://www.ijert.org/online-kannada-handwritten-characters-and-numerical-recognition-using-cnn-classifier
Upadhye GD, Kulkarni UV, Mane DT. Improved Model Configuration Strategies for Kannada Handwritten Numeral Recognition. Image Analysis & Stereology. 2021;40(3):181–191. Available from: https://doi.org/10.5566/ias.2586
Puri S, Singh SP. Toward recognition and classification of Hindi handwritten document imag. Ambient Communications and Computer Systems. In: Advances in Intelligent Systems and Computing. (Vol. 904, pp. 497-507) Springer. 2019.
Thakur A, Kaur A. Devanagari Handwritten Character Recognition Using Neural Network. International Journal of Scientific & Technology Research. 2019;8(10):1736–1744. Available from: http://www.ijstr.org/final-print/oct2019/Devanagari-Handwritten-Character-Recognition-Using-Neural-Network.pdf
Jangid M, Srivastava S. Handwritten Devanagari Character Recognition Using Layer-Wise Training of Deep Convolutional Neural Networks and Adaptive Gradient Methods. Journal of Imaging. 2018;4(2):41. Available from: https://doi.org/10.3390/jimaging4020041
Sahare P, Dhok SB. Multilingual Character Segmentation and Recognition Schemes for Indian Document Images. IEEE Access. 2018;6:10603–10617. Available from: https://doi.org/10.1109/ACCESS.2018.2795104
Baranidharana B, Kandpal A, Chakravorty A. Hindi Handwritten Character Recognition using CNN. International Journal of Advanced Science and Technology. 2020;29(06):58–66. Available from: http://sersc.org/journals/index.php/IJAST/article/view/11294
Aneja N, Aneja S. Transfer Learning using CNN for Handwritten Devanagari Character Recognition. 1st International Conference on Advances in Information Technology (ICAIT). 2019;p. 293–296. Available from: https://doi.org/10.1109/ICAIT47043.2019.8987286
Prashanth DS, Mehta RVK, Ramana K, Bhaskar V. Handwritten Devanagari Character Recognition Using Modified Lenet and Alexnet Convolution Neural Networks. Wireless Personal Communications. 2022;122(1):349–378. Available from: https://doi.org/10.1007/s11277-021-08903-4
Prashanth DS, Panini CN. KNN classification of Kannada Characters using Hu’s Seven Variants and Zernike Moment. 2017. Available from: https://www.researchgate.net/publication/312495850
Kaur S, Sagar BB. Brahmi character recognition based on SVM (support vector machine) classifier using image gradient features. Journal of Discrete Mathematical Sciences and Cryptography. 2019;22(8):1365–1381. Available from: https://doi.org/10.1080/09720529.2019.1692445
Parekh KA, Goswami MM, Mitra SK. Handwritten Numeral Recognition Using Polar Histogram of Low-Level Stroke Features. Proceedings of 3rd International Conference on Computer Vision and Image Processing. 2020;p. 169–181. Available from: https://doi.org/10.1007/978-981-32-9088-4_15
Parikshith H, Rajath SMN, Shwetha D, Sindhu CM, Ravi P. Handwritten Character Recognition of Kannada Language Using Convolutional Neural Networks and Transfer Learning. IOP Conference Series: Materials Science and Engineering. 2021;1110(1):012003. Available from: https://doi.org/10.1088/1757-899X/1110/1/012003
Bhattacharya N, Roy PP, Pal U, Setua SK. Online Bangla handwritten word recognition. Malaysian Journal of Computer Science. 2018;31(4):300–310. Available from: https://doi.org/10.22452/mjcs.vol31no4.4
Singh S, Chauhan VK, Smith EHB. A self controlled RDP approach for feature extraction in online handwriting recognition using deep learning. Applied Intelligence. 2020;50(7):2093–2104. Available from: https://doi.org/10.1007/s10489-020-01632-4
Ghosh R, Vamshi C, Kumar P. RNN based online handwritten word recognition in Devanagari and Bengali scripts using horizontal zoning. Pattern Recognition. 2019;92:203–218. Available from: https://doi.org/10.1016/j.patcog.2019.03.030

Copyright

© 2023 Siddanna & Kiran. This is an open-access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited. Published By Indian Society for Education and Environment (iSee)