A Novel Script-Rule-Based Character Segmentation Method for Devanagari Script in a Natural Scene Image
Keywords:
Character Segmentation, CapsNet, Devanagari Script, DBN, Natural Scene Image, Naïve BayesAbstract
Devanagari script is one of the most popular and commonly utilized scripts across India. In this script, every character signifies a consonant sound along with an integral vowel sound. During the last decade, huge research is conducted for text localization and detection of various languages such as Japanese, English, Chinese, and others in scene images. However, there is limited research studies were performed on character segmentation for Devanagari Script. The existing methods used in the Devanagari script have some limits such as low accuracy in line, word, and character segmentation, dependency on handcrafted features in certain settings, computational complexity, and segmentation errors, etc. This research proposes a novel script-rule-based character segmentation model for the Devanagari script in a natural scene image setting. This proposed method is based on Deep Belief Network (DBN), Capsule Neural Network (CapsNet), and Naïve Bayes. Additionally, to recognize the middle region of the segmented region of the image, a modified recurrent neural network (RNN) model has been utilized. The measured performance metrics such as accuracy, precision, recall, and F1-score on Vpd Datasets using AdaDelta optimizer are 98.62%, 98,12%, 97.43%, and 97.01%, respectively. It is found that the proposed script rule-based character segmentation method obtains very optimized results.
Downloads
References
D. Cao, Y. Zhong, L. Wang, Y. He, and J. Dang, “Scene Text Detection in Natural Images: A Review,” Symmetry (Basel)., vol. 12, no. 12, p. 1956, Nov. 2020, doi: 10.3390/sym12121956.
V. Naosekpam and N. Sahu, “Text detection, recognition, and script identification in natural scene images: A Review,” Int. J. Multimed. Inf. Retr., 2022, doi: 10.1007/s13735-022-00243-8.
N. Gupta and A. S. Jalal, “Traditional to transfer learning progression on scene text detection and recognition: A survey,” Artif. Intell. Rev., 2022, doi: 10.1007/s10462-021-10091-3.
T. Khan, R. Sarkar, and A. F. Mollah, “Deep learning approaches to scene text detection: a comprehensive review,” Artif. Intell. Rev., 2021, doi: 10.1007/s10462-020-09930-6.
A. Khalil, M. Jarrah, M. Al-Ayyoub, and Y. Jararweh, “Text detection and script identification in natural scene images using deep learning,” Comput. Electr. Eng., 2021, doi: 10.1016/j.
compeleceng.2021.107043.
M. Ibrayim, Y. Li, and A. Hamdulla, “Scene Text Detection Based on Two-Branch Feature Extraction,” Sensors, 2022, doi: 10.3390/s22166262.
D. Pandey, B. K. Pandey, and S. Wairya, “Hybrid deep neural network with adaptive galactic swarm optimization for text extraction from scene images,” Soft Comput., 2021, doi: 10.1007/s00500-020-05245-4.
K. Boukthir, A. M. Qahtani, O. Almutiry, H. Dhahri, and A. M. Alimi, “Reduced annotation based on deep active learning for arabic text detection in natural scene images,” Pattern Recognit. Lett., 2022, doi: 10.1016/j.patrec.2022.03.016.
D. Zhong, P. Shivakumara, L. Nandanwar, U. Pal, M. Blumenstein, and Y. Lu, “Local Resultant Gradient Vector Difference and Inpainting for 3D Text Detection in the Wild,” Int. J. Pattern Recognit. Artif. Intell., 2022, doi: 10.1142/S0218001422530056.
A. Ali, M. Pickering, and K. Shafi, “Urdu Natural Scene Character Recognition using Convolutional Neural Networks,” 2018. doi: 10.1109/ASAR.2018.8480202.
O. Y. Ling, L. B. Theng, A. C. Weiyen, and M. Christopher, “Development of Vertical Text Interpreter for Natural Scene Images,” IEEE Access, 2021, doi: 10.1109/ACCESS.2021.
A. A. A. Ali and S. Mallaiah, “Intelligent handwritten recognition using hybrid CNN architectures based-SVM classifier with dropout,” J. King Saud Univ. - Comput. Inf. Sci., vol. 34, no. 6, pp. 3294–3300, 2022, doi: 10.1016/j.jksuci.2021.01.012.
M. S. V. Dharsini, N. Kamath, Y. Chaudhary, A. Sriramoju, and N. Anurag, “Devanagri Character Image Recognition and Conversion into Text using Long Short Term Memory,” 2022. doi: 10.1109/ICDSAAI55433.2022.10028963.
S. Ahlawat, A. Choudhary, A. Nayyar, S. Singh, and B. Yoon, “Improved handwritten digit recognition using convolutional neural networks (Cnn),” Sensors (Switzerland), 2020, doi: 10.3390/s20123344.
B. Al-Sadawi, A. Hussain, and N. S. Ali, “High-Performance Printed Arabic Optical Character Recognition System Using ANN Classifier,” 2021. doi: 10.1109/PICICT53635.2021.00013.
H. Butt, M. R. Raza, M. J. Ramzan, M. J. Ali, and M. Haris, “Attention-Based CNN-RNN Arabic Text Recognition from Natural Scene Images,” Forecasting, 2021, doi: 10.3390/forecast3030033.
M. Sonkusare, R. Gupta, and A. Moghe, “A Review on Character Segmentation Approach for Devanagari Script,” 2021. doi: 10.1007/978-981-16-2248-9_19.
S. Gunna, R. Saluja, and C. V. Jawahar, “Improving Scene Text Recognition for Indian Languages with Transfer Learning and Font Diversity,” J. Imaging, 2022, doi: 10.3390/jimaging8040086.
S. Bin Ahmed, S. Naz, M. I. Razzak, and R. Yusof, “Arabic cursive text recognition from natural scene images,” Appl. Sci., 2019, doi: 10.3390/app9020236.
S. Bin Ahmed, S. Naz, M. I. Razzak, and R. B. Yusof, “A Novel Dataset for English-Arabic Scene Text Recognition (EASTR)-42K and Its Evaluation Using Invariant Feature Extraction on Detected Extremal Regions,” IEEE Access, 2019, doi: 10.1109/ACCESS.
2895876.
S. Y. Arafat and M. J. Iqbal, “Urdu-Text Detection and Recognition in Natural Scene Images Using Deep Learning,” IEEE Access, 2020, doi: 10.1109/ACCESS.2020.2994214.
H. Raj and R. Ghosh, “Devanagari text extraction from natural scene images,” in 2014 International Conference on Advances in Computing, Communications and Informatics (ICACCI), Sep. 2014, pp. 513–517. doi: 10.1109/ICACCI.2014.6968472.
M. Jangid and S. Srivastava, “Handwritten Devanagari character recognition using layer-wise training of deep convolutional neural networks and adaptive gradient methods,” J. Imaging, 2018, doi: 10.3390/jimaging4020041.
A. Moudgil, S. Singh, V. Gautam, S. Rani, and S. H. Shah, “Handwritten devanagari manuscript characters recognition using capsnet,” Int. J. Cogn. Comput. Eng., 2023, doi: 10.1016/j.ijcce.
02.001.
S. R. Narang, M. Kumar, and M. K. Jindal, “DeepNetDevanagari: a deep learning model for Devanagari ancient character recognition,” Multimed. Tools Appl., 2021, doi: 10.1007/s11042-021-10775-6.
Downloads
Published
How to Cite
Issue
Section
License
This work is licensed under a Creative Commons Attribution-ShareAlike 4.0 International License.
All papers should be submitted electronically. All submitted manuscripts must be original work that is not under submission at another journal or under consideration for publication in another form, such as a monograph or chapter of a book. Authors of submitted papers are obligated not to submit their paper for publication elsewhere until an editorial decision is rendered on their submission. Further, authors of accepted papers are prohibited from publishing the results in other publications that appear before the paper is published in the Journal unless they receive approval for doing so from the Editor-In-Chief.
IJISAE open access articles are licensed under a Creative Commons Attribution-ShareAlike 4.0 International License. This license lets the audience to give appropriate credit, provide a link to the license, and indicate if changes were made and if they remix, transform, or build upon the material, they must distribute contributions under the same license as the original.
