A Hybrid CNN-RNN Model for Automated Recognition of Kannada Characters in Ancient Inscriptions
Received: 2 August 2024 | Revised: 19 September 2024 | Accepted: 22 September 2024 | Online: 30 October 2024
Corresponding author: B. K. Rajithkumar
Abstract
This study presents a novel approach for the automated recognition of Kannada characters in ancient inscriptions using a hybrid Convolutional Neural Network and Recurrent Neural Network (CNN-RNN) model. The unique features of stone inscriptions, such as erosion, uneven surfaces, and varying font styles, pose significant challenges to traditional character recognition systems. The proposed hybrid model leverages the strengths of CNNs for feature extraction and RNNs for sequence prediction, enabling robust recognition of complex and degraded characters. The proposed model was trained and tested on a curated dataset of annotated Kannada inscriptions, achieving an impressive accuracy of 95%. This high accuracy demonstrates the model's effectiveness in deciphering ancient scripts, which is critical for the preservation and study of historical texts. The results highlight the potential of deep learning techniques in advancing the field of epigraphy and cultural heritage preservation.
Keywords:
kannada character recognition, image processing, epigraphyDownloads
References
B. K. Rajithkumar, B. V. Uma, and H. S. Mohana, "Recognition of Historical Kannada Manuscripts using Convolution Neural Network," International Journal of Intelligent Systems and Applications in Engineering, vol. 12, no. 3, pp. 1138–1147, Mar. 2024.
B. K. RajithKumar and H. S. Mohana, "Template Matching Method for Recognition of Stone Inscripted Kannada Characters of Different Time Frames Based on Correlation Analysis," International Journal of Electrical and Computer Engineering, vol. 4, no. 5, pp. 719–729, Oct. 2014. DOI: https://doi.org/10.11591/ijece.v4i5.6333
T. N. Sridevi, et al., "Deep Convolution Neural Network for Degraded Printed Kannada Character Recognitions," Indian Journal of Computer Science and Engineering, vol. 4, no. 3, May 2021.
H. S. Mohana, B. K. Rajithkumar, B. R. Sujatha, and K. Navya, "Extraction of stone in-scripted Kannada characters using sift algorithm based image mosaic," International Journal of Electronics & Communication Technology, vol. 5, no. 2, 2014.
H. S. Mohana and B. K. Rajithkumar, "Era identification and recognition of Ganga and Hoysala phase Kannada stone inscriptions characters using advance recognition algorithm," in 2014 International Conference on Control, Instrumentation, Communication and Computational Technologies (ICCICCT), Kanyakumari District, India, Jul. 2014, pp. 659–665. DOI: https://doi.org/10.1109/ICCICCT.2014.6993043
M. Bhalekar and M. Bedekar, "D-CNN: A New model for Generating Image Captions with Text Extraction Using Deep Learning for Visually Challenged Individuals," Engineering, Technology & Applied Science Research, vol. 12, no. 2, pp. 8366–8373, Apr. 2022. DOI: https://doi.org/10.48084/etasr.4772
S. R. Basha, J. K. Rani, and J. J. C. P. Yadav, "A Novel Summarization-based Approach for Feature Reduction Enhancing Text Classification Accuracy," Engineering, Technology & Applied Science Research, vol. 9, no. 6, pp. 5001–5005, Dec. 2019. DOI: https://doi.org/10.48084/etasr.3173
R. Prabavathi and R. Karthikeyan, "Eighth Century Tamil Character Recognition," International Journal of Innovative Research in Computer and Communication Engineering, vol. 3, no. 2, pp. 707–712, 2015.
H. T. Chandrakala and G. Thippeswamy, "Deep Convolutional Neural Networks for Recognition of Historical Handwritten Kannada Characters," in Frontiers in Intelligent Computing: Theory and Applications, 2020, pp. 69–77. DOI: https://doi.org/10.1007/978-981-13-9920-6_7
F. Lombardi and S. Marinai, "Deep Learning for Historical Document Analysis and Recognition—A Survey," Journal of Imaging, vol. 6, no. 10, Oct. 2020, Art. no. 110. DOI: https://doi.org/10.3390/jimaging6100110
A. S. Rao, S. Nayak, A. K. Arpitha, C. Nayak, and M .S. Nayak, "Exploring Deep Learning Techniques for Kannada Handwritten Character Recognition: A Boon for Digitization," International Journal of Advanced Science and Technology, vol. 29, no. 5, pp. 11078–11093, 2020.
H. T. Chandrakala and G. Thippeswamy, "A comprehensive survey on OCR techniques for kannada script," International Journal of Scientific Research, vol. 5, no. 4, pp. 35–39, 2016. DOI: https://doi.org/10.21275/v5i4.NOV162366
M. R. Murty, J. V. R. Murthy, and P. V. G. D. Prasad Reddy, "Text document classification based-on least square support vector machines with singular value decomposition," International Journal of Computer Applications, vol. 27, no. 7, pp. 21–26, 2011. DOI: https://doi.org/10.5120/3312-4540
S. Roy, N. Das, M. Kundu, and M. Nasipuri, "Handwritten isolated Bangla compound character recognition: A new benchmark using a novel deep learning approach," Pattern Recognition Letters, vol. 90, pp. 15–21, Apr. 2017. DOI: https://doi.org/10.1016/j.patrec.2017.03.004
J. Jomy, K. Balakrishnan, and K. V. Pramod, "A System for Offline Recognition of Handwritten Characters in Malayalam Script," International Journal of Image, Graphics and Signal Processing, vol. 5, no. 4, pp. 53–59, Apr. 2013. DOI: https://doi.org/10.5815/ijigsp.2013.04.07
A. B. Shinde and Y. H. Dandawate, "Shirorekha extraction in Character Segmentation for printed devanagri text in Document Image Processing," in 2014 Annual IEEE India Conference (INDICON), Pune, India, Dec. 2014, pp. 1–7. DOI: https://doi.org/10.1109/INDICON.2014.7030535
S. Sharma, A. Sasi, and A. N. Cheeran, "A SVM based character recognition system," in 2017 2nd IEEE International Conference on Recent Trends in Electronics, Information & Communication Technology (RTEICT), Bangalore, India, May 2017, pp. 1703–1707. DOI: https://doi.org/10.1109/RTEICT.2017.8256890
D. K. Gowda and V. Kanchana, "Kannada Handwritten Character Recognition and Classification Through OCR Using Hybrid Machine Learning Techniques," in 2022 IEEE International Conference on Data Science and Information System (ICDSIS), Hassan, India, Jul. 2022, pp. 1–6. DOI: https://doi.org/10.1109/ICDSIS55133.2022.9915906
S. Puri and S. P. Singh, "An efficient Devanagari character classification in printed and handwritten documents using SVM," Procedia Computer Science, vol. 152, pp. 111–121, Jan. 2019. DOI: https://doi.org/10.1016/j.procs.2019.05.033
N. Sarika, N. Sirisala, and M. S. Velpuru, "CNN based Optical Character Recognition and Applications," in 2021 6th International Conference on Inventive Computation Technologies (ICICT), Coimbatore, India, Jan. 2021, pp. 666–672. DOI: https://doi.org/10.1109/ICICT50816.2021.9358735
N. Sandhya and R. Krishnan, "Broken kannada character recognition — A neural network based approach," in 2016 International Conference on Electrical, Electronics, and Optimization Techniques (ICEEOT), Chennai, India, Mar. 2016, pp. 2047–2050. DOI: https://doi.org/10.1109/ICEEOT.2016.7755047
N. Sandhya, R. Krishnan, D. R. R. Babu, and N. B. Rao, "An efficient approach for handling degradation in character recognition," International Journal of Advanced Intelligence Paradigms, vol. 14, no. 1–2, pp. 14–29, Jan. 2019. DOI: https://doi.org/10.1504/IJAIP.2019.102960
V. N. M. Aradhya, G. H. Kumar, S. Noushath, and P. Shivakumara, "Fisher Linear Discriminant Analysis based Technique Useful for Efficient Character Recognition," in 2006 Fourth International Conference on Intelligent Sensing and Information Processing, Bangalore, India, Dec. 2006, pp. 49–52. DOI: https://doi.org/10.1109/ICISIP.2006.4286060
N. Sandhya, R. Krishnan, and D. R. Ramesh Babu, "A novel local enhancement technique for rebuilding Broken characters in a degraded Kannada script," in 2015 IEEE International Advance Computing Conference (IACC), Banglore, Jun. 2015, pp. 176–179. DOI: https://doi.org/10.1109/IADCC.2015.7154693
Downloads
How to Cite
License
Copyright (c) 2024 B. K. Rajithkumar, B. V. Uma, H. S. Mohana
This work is licensed under a Creative Commons Attribution 4.0 International License.
Authors who publish with this journal agree to the following terms:
- Authors retain the copyright and grant the journal the right of first publication with the work simultaneously licensed under a Creative Commons Attribution License that allows others to share the work with an acknowledgement of the work's authorship and initial publication in this journal.
- Authors are able to enter into separate, additional contractual arrangements for the non-exclusive distribution of the journal's published version of the work (e.g., post it to an institutional repository or publish it in a book), with an acknowledgement of its initial publication in this journal.
- Authors are permitted and encouraged to post their work online (e.g., in institutional repositories or on their website) after its publication in ETASR with an acknowledgement of its initial publication in this journal.
