Smart Network-Based Classification of Handwritten Basic, Modified, and Complex Conjunct Characters in Devanagari Script
| International Journal of Electronics and Communication Engineering |
| © 2025 by SSRG - IJECE Journal |
| Volume 12 Issue 11 |
| Year of Publication : 2025 |
| Authors : Pallavi Patil, Kevin Noronha |
10.14445/23488549/IJECE-V12I11P117
How to Cite?
Pallavi Patil, Kevin Noronha, "Smart Network-Based Classification of Handwritten Basic, Modified, and Complex Conjunct Characters in Devanagari Script," SSRG International Journal of Electronics and Communication Engineering, vol. 12, no. 11, pp. 201-215, 2025. Crossref, https://doi.org/10.14445/23488549/IJECE-V12I11P117
Abstract:
The Devanagari script is extensively used for documenting information in both the government and private sectors in India. In Devanagari documents, modified and conjunct characters frequently appear alongside consonants. Intricate forms of these characters, combined with variations in writing patterns, make the recognition process a challenging task. Most Devanagari datasets available online lack modified and conjunct characters. To address this issue, a dataset comprising basic, modified, and complex conjunct characters of the Devanagari script, which includes 580 classes, is created. Additionally, a segmentation algorithm is developed to automatically segment filled forms from different writers, thereby accelerating dataset processing. To classify these characters accurately, three convolutional neural networks, SmartNet1, SmartNet2, and SmartNet3, were designed by experimenting with various hyperparameters, such as the number of layers, filters, nodes in the fully connected layer, and kernel size. Each SmartNet was modified using a support vector machine and a k-nearest neighbors classifier, yielding a total of nine configurations. All nine configurations were tested on both self-generated and benchmark datasets. These networks are relatively shallow and have fewer parameters, enabling faster convergence and accomplishing remarkable results across different datasets.
Keywords:
Conjunct, Devanagari, Modified, Segmentation, SFIT_Char.
References:
[1] Somnath Chatterjee et al., “City Name Recognition for Indian Postal Automation: Exploring Script Dependent and Independent Approach,” Multimedia Tools and Applications, vol. 83, pp. 22371-22394, 2023.
[CrossRef] [Google Scholar] [Publisher Link]
[2] Ujjwal Bhattacharya, and B.B. Chaudhuri, “Handwritten Numeral Databases of Indian Scripts and Multistage Recognition of Mixed Numerals,” IEEE Transactions on Pattern Analysis and Machine Intelligence, vol. 31, pp. 444-457, 2009.
[CrossRef] [Google Scholar] [Publisher Link]
[3] Jamshed Memon et al., “Handwritten Optical Character Recognition (HOCR): A Comprehensive Systematic Literature Review,” IEEE Access, vol. 8, pp. 142642-142668, 2020.
[CrossRef] [Google Scholar] [Publisher Link]
[4] Ismet Zeki Yalniz, and R. Manmatha, “Dependence Models for Searching Text in Document Images,” IEEE Transactions on Pattern Analysis and Machine Intelligence, vol. 41, no. 1, pp. 49-63, 2019.
[CrossRef] [Google Scholar] [Publisher Link]
[5] Chetan Ralekar et al., “Intelligent Identification of Ornamental Devanagari Characters Inspired by Visual Fixations,” 2019 International Conference on Document Analysis and Recognition Workshops (ICDARW), Sydney, NSW, Australia, pp. 14-19, 2019.
[CrossRef] [Google Scholar] [Publisher Link]
[6] Diptee Vishwanath Chikmurge, and Shriram Raghunathan, “Enhancing Marathi Handwritten Character Recognition Using Ensemble Learning,” Signal Processing, vol. 40, no. 1, pp. 327-334, 2023.
[CrossRef] [Google Scholar] [Publisher Link]
[7] Sandeep Dwarkanath Pande et al., “Digitization of Handwritten Devanagari Text using CNN Transfer Learning – A Better Customer Service Support,” Neuroscience Informatics, vol. 2, no. 3, 2022.
[CrossRef] [Google Scholar] [Publisher Link]
[8] Srinidhi Karthikeyan et al., “An OCR Post- Correction Approach Using Deep Learning for Processing Medical Reports,” IEEE Transactions on Circuits and Systems for Video Technology, vol. 32, no. 5, pp. 2574-2581, 2022.
[CrossRef] [Google Scholar] [Publisher Link]
[9] Hubert et al., “Classifying Promotion Images Using Optical Character Recognition and Naïve Bayes Classifier,” Procedia Computer Science, vol. 179, pp. 498-506, 2021.
[CrossRef] [Google Scholar] [Publisher Link]
[10] Senka Drobac, and Krister Lindén, “Optical Character Recognition with Neural Networks and Post-Correction with Finite State Methods,” International Journal on Document Analysis and Recognition, vol. 23, pp. 279-295, 2020.
[CrossRef] [Google Scholar] [Publisher Link]
[11] Noor Masrur et al., “Dissimilarity Matrix: Bridging the Gap of Bangla OCR Error Correction through a Novel Approach,” 2023 5th International Conference on Sustainable Technologies for Industry 5.0 (STI), Dhaka, Bangladesh, pp. 1-6, 2023.
[CrossRef] [Google Scholar] [Publisher Link]
[12] Zhenrong Zhang et al., “Bengali Handwritten Grapheme Recognition Using Cutmix-Based Data Augmentation,” Proceedings of the 3rd World Symposium on Software Engineering, pp. 145-149, 2021.
[CrossRef] [Google Scholar] [Publisher Link]
[13] Minesh Mathew, Mohit Jain, and C.V. Jawahar, “Benchmarking Scene Text Recognition in Devanagari, Telugu and Malayalam,” 2017 14th IAPR International Conference on Document Analysis and Recognition (ICDAR), Kyoto, Japan, pp. 42-46, 2017.
[CrossRef] [Google Scholar] [Publisher Link]
[14] Vrushali. T. Lanjewar, and R.N. Khobragade, “Transfer Learning using Pre-Trained AlexNet for Marathi Handwritten Compound Character Image Classification,” 2021 International Conference on Intelligent Technologies (CONIT), Hubli, India, pp. 1-7, 2021.
[CrossRef] [Google Scholar] [Publisher Link]
[15] Madhuri Yadav, Ravindra Kumar Purwar, and Mamta Mittal, “Handwritten Hindi Character Recognition: A Review,” IET Image Processing, vol. 12, no. 11, pp. 1919-1933, 2018.
[CrossRef] [Google Scholar] [Publisher Link]
[16] Meduri Avadesh, and Navneet Goyal, “Optical Character Recognition for Sanskrit Using Convolution Neural Networks,” 2018 13th IAPR International Workshop on Document Analysis Systems (DAS), Vienna, Austria, pp. 447-452, 2018.
[CrossRef] [Google Scholar] [Publisher Link]
[17] Arti Khaparde, Vaidehi Deshmukh, and Manisha Kowdiki, “Enhanced Nature-Inspired Algorithm-Based Hybrid Deep Learning for Character Recognition in Sanskrit Language,” Sensing and Imaging, vol. 24, 2023.
[CrossRef] [Google Scholar] [Publisher Link]
[18] Dhruv Kudale et al., “Textron: Weakly Supervised Multilingual Text Detection through Data Programming,” Proceedings of the IEEE/CVF Winter Conference on Applications of Computer Vision (WACV), pp. 2871-2880, 2024.
[Google Scholar] [Publisher Link]
[19] Parul Sahare, and Sanjay B. Dhok, “Multilingual Character Segmentation and Recognition Schemes for Indian Document Images,” IEEE Access, vol. 6, pp. 10603-10617, 2018.
[CrossRef] [Google Scholar] [Publisher Link]
[20] Sonika Rani Narang, M.K. Jindal, and Munish Kumar, “Ancient Text Recognition: A Review,” Artificial Intelligence Review, vol. 53, pp. 5517-5558, 2020.
[CrossRef] [Google Scholar] [Publisher Link]
[21] Amar Jindal, and Rajib Ghosh, “A Hybrid Deep Learning Model to Recognize Handwritten Characters in Ancient Documents in Devanagari and Maithili Scripts,” Multimedia Tools and Applications, vol. 83, pp. 8389-8412, 2024.
[CrossRef] [Google Scholar] [Publisher Link]
[22] Anuradha Choudry, Intermediate Level of Spoken Sanskrit, Swayam - NPTEL, 2018. [Online] Available: https://onlinecourses.nptel.ac.in/noc19_hs53/preview
[23] Partha Pratim Roy et al., “HMM-based Indic Handwritten Word Recognition using Zone Segmentation,” Pattern Recognition, vol. 60, pp. 1057-1075, 2016.
[CrossRef] [Google Scholar] [Publisher Link]
[24] Mahesh Jangid, and Sumit Srivastava, “Handwritten Devanagari Character Recognition Using Layer-wise Training of Deep Convolutional Neural Networks and Adaptive Gradient Methods,” Journal of Imaging, vol. 4, no. 2, pp. 1-14, 2018. [CrossRef] [Google Scholar] [Publisher Link]
[25] Shrawan Ram, Shloak Gupta, and Basant Agarwal, “Devanagri Character Recognition Model using Deep Convolution Neural Network,” Journal of Statistics and Management Systems, vol. 21, no. 4, pp. 593-599, 2018.
[CrossRef] [Google Scholar] [Publisher Link]
[26] Sonika Narang, M.K Jindal, and Munish Kumar, “Devanagari Ancient Documents Recognition Using Statistical Feature Extraction Techniques,” Sādhanā, vol. 44, 2019.
[CrossRef] [Google Scholar] [Publisher Link]
[27] Shalaka Prasad Deore, and Albert Pravin, “Devanagari Handwritten Character Recognition using Fine- Tuned Deep Convolutional Neural Network on Trivial Dataset,” Sādhanā, vol. 45, 2020.
[CrossRef] [Google Scholar] [Publisher Link]
[28] Duddela Sai Prashanth et al., “Handwritten Devanagari Character Recognition Using Modified Lenet and Alexnet Convolution Neural Networks,” Wireless Personal Communications, vol. 122, pp. 349-378, 2022.
[CrossRef] [Google Scholar] [Publisher Link]
[29] Snehal S. Gaikwad, S.L Nalbalwar, and A.B. Nandgaonkar, “Devanagari Handwritten Characters Recognition using DCT, Geometric and Hue Moments Feature Extraction Techniques,” Sādhanā, vol. 47, 2022.
[CrossRef] [Google Scholar] [Publisher Link]
[30] Danveer Rajpal, and Akhil Ranjan Garg, “Deep Learning Model for Recognition of Handwritten Devanagari Numerals with Low Computational Complexity and Space Requirements,” IEEE Access, vol. 11, pp. 49530-49539, 2023.
[CrossRef] [Google Scholar] [Publisher Link]
[31] Khushi Sinha, and Richa Gupta, “Enhancing Handwritten Devanagari Character Recognition via GAN-Generated Synthetic Data,” Proceedings of the 2024 Sixteenth International Conference on Contemporary Computing, pp. 484-489, 2024.
[CrossRef] [Google Scholar] [Publisher Link]
[32] Chetan Sharma et al., “Advancements in Handwritten Devanagari Character Recognition: A Study on Transfer Learning and VGG16 Algorithm,” Discover Applied Sciences, vol. 6, pp. 1-15, 2024.
[CrossRef] [Google Scholar] [Publisher Link]
[33] Bharati Yadav, Ajay Indian, and Chandra Prakash Gupta, “Offline Handwritten Devanagari Character Recognition Using Modified Capsule Neural Network,” Procedia Computer Science, vol. 258, pp. 2878-2887, 2025.
[CrossRef] [Google Scholar] [Publisher Link]
[34] Vikas J. Dongre, and Vijay H. Mankar, “Development of Comprehensive Devnagari Numeral and Character Database for Offline Handwritten Character Recognition,” Applied Computational Intelligence and Soft Computing, vol. 2012, no. 1, pp. 1-5, 2012.
[CrossRef] [Google Scholar] [Publisher Link]
[35] R. Kumar, A. Kumar, and P. Ahmed, “A Benchmark Dataset for Devnagari Document Recognition,” Recent Advances in Electrical Engineering Series, pp. 258-263, 2013.
[Google Scholar]
[36] Shailesh Acharya, Ashok Kumar Pant, and Prashnna Kumar Gyawali, “Deep Learning based Large Scale Handwritten Devanagari Character Recognition,” 2015 9th International Conference on Software, Knowledge, Information Management and Applications (SKIMA), Kathmandu, Nepal, pp. 1-6, 2015.
[CrossRef] [Google Scholar] [Publisher Link]
[37] Anuj Bhardwaj, and Ravendra Singh, “Handwritten Devanagari Character Recognition Using Deep Learning - Convolutional Neural Network (CNN) Model,” PalArch’s Journal of Archaeology of Egypt / Egyptology, vol. 17, no. 6, pp. 7965-7984, 2020.
[Google Scholar] [Publisher Link]