The week's pick
Reseach Article
Hybrid Approach for Recognition of Isolated Handwritten Fraction Notations in Telugu Script
| International Journal of Computer Applications |
| Foundation of Computer Science (FCS), NY, USA |
| Volume 186 - Number 19 |
| Year of Publication: 2024 |
| Authors: Vempati Lakshmi Sravani, Piyush Pratap Singh |
10.5120/ijca2024923600
|
Vempati Lakshmi Sravani, Piyush Pratap Singh . Hybrid Approach for Recognition of Isolated Handwritten Fraction Notations in Telugu Script. International Journal of Computer Applications. 186, 19 ( May 2024), 38-45. DOI=10.5120/ijca2024923600
Abstract
Recognition of handwritten digits in regional Indian languages presents formidable challenges due to the vast array of scripts and variability in writing styles. Despite notable advancements in research techniques and databases for recognizing "0 to 9" digits in Indian scripts such as Bangla, Kannada, Devanagari, Oriya, and Telugu, there exists a conspicuous gap in research explicitly addressing the recognition of "fraction" notations unique to the Telugu script. Consequently, the proposed methodology integrates deep learning techniques, employing Convolutional Neural Networks (CNNs) for feature extraction and Support Vector Machines (SVMs) with various kernels for classification. This approach is explicitly tailored towards recognizing handwritten Telugu "fraction" notations, aiming to fill the existing research void in this domain. To facilitate model training, a comprehensive dataset is curated comprising 4000 handwritten Telugu fraction images covering eight distinct notation classes and enhanced dataset diversity through data augmentation techniques. Extensive experimental validation showcases the efficacy of the proposed hybrid CNN-SVM framework, achieving an impressive accuracy of 99.86% using the RBF kernel, outperforming the standalone CNN model (99.67% accuracy). These findings highlight the effectiveness of the proposed method in this underexplored field of recognizing handwritten Telugu fractions that can contribute to the digital preservation of ancient Telugu manuscripts and fostering linguistic diversity, paving the way for broader language technology applications.
References
- Charles Philip Brown, 1856 (2006), “The Grammar of the Telugu Language, 2nd Edition,” Chennai: Asian Educational Society, ISBN 81-206-0041-X.
- Alexander Duncan Campbell, 1849 (1991), “Grammar of the Teloogoo Language, 3rd Edition,” Chennai: Asian Educational Society, ISBN 81-206-0366-4.
- The Unicode Consortium. The Unicode Standard, Version 15.1.0, (South San Francisco, CA: The Unicode Consortium, 2023, ISBN 978-1-936213-33-7)
- Charles Philip Brown, Revised by M Venkata Ratnam, W H Campbell, Kandukuri Viresalingam. 1903 (2004), “Telugu-English Dictionary = Nighantuvu Telugu - Inglish, 2nd Edition,” Chennai: Asian Educational Society, ISBN 81-206-0037-1.
- Puduru Seetarama Sastry, 1847 (1916), “Peddabālaśiksha,” Cennapuri: Vavilla Ramaswami Sastrulu and Sons, Chennai.
- Nāgārjuna Venna. "Telugu Measures and Arithmetic Marks," JTC1/SC2/WG2 N3156, International Organization for Standardization.
- U. Bhattacharya and B. B. Chaudhuri, "Handwritten Numeral Databases of Indian Scripts and Multistage Recognition of Mixed Numerals," in IEEE Transactions on Pattern Analysis and Machine Intelligence, vol. 31, no. 3, pp. 444–457, March 2009, doi: 10.1109/TPAMI.2008.88.
- U. Pal, N. Sharma, T. Wakabayashi and F. Kimura, "Handwritten Numeral Recognition of Six Popular Indian Scripts," Ninth International Conference on Document Analysis and Recognition (ICDAR 2007), Curitiba, Brazil, 2007, pp. 749–753, doi: 10.1109/ICDAR.2007.4377015.
- Nibaran Das, Jagan Mohan Reddy, Ram Sarkar, Subhadip Basu, Mahantapas Kundu, Mita Nasipuri, Dipak Kumar Basu, “A statistical–topological feature combination for recognition of handwritten numerals,” Applied Soft Computing, Volume 12, Issue 8, 2012, pp. 2486-2495, doi: 10.1016/j.asoc.2012.03.039.
- K. Roy, T. Pal, U. Pal, and F. Kimura, "Oriya handwritten numeral recognition system," Eighth International Conference on Document Analysis and Recognition (ICDAR'05), Seoul, Korea (South), 2005, pp. 770-774 Vol. 2, doi: 10.1109/ICDAR.2005.183.
- Shailedra Kumar Shrivastava, Sanjay S. Gharde, “Support Vector Machine for Handwritten Devanagari Numeral Recognition,” International Journal of Computer Applications. 7, 11 (October 2010), 9–14, doi: 10.5120/1293-1769.
- B.V.Dhandra, R.G.Benne, Mallikarjun Hangarge, “Kannada, Telugu and Devanagari Handwritten Numeral Recognition with Probabilistic Neural Network: A Novel Approach,” Recent Trends in Image Processing and Pattern Recognition. RTIPPR, 2 (None 2010), 83-88.
- Stuti Asthana, Farha Haneef, and Rakesh K. Bhujade, "Handwritten multiscript numeral recognition using artificial neural networks," International Journal of Soft Computing and Engineering, 1.1 (2011), 1–5.
- Adarsh Trivedi, Siddhant Srivastava, Apoorva Mishra, Anupam Shukla, and Ritu Tiwari, "Hybrid evolutionary approach for Devanagari handwritten numeral recognition using Convolutional Neural Network," Procedia Computer Science, 125 (2018): 525-532.
- Nobuyuki Otsu (1979), "A threshold selection method from gray-level histograms," IEEE Trans. Sys., Man., Cyber. 9 (1): 62–66, doi:10.1109/TSMC.1979.4310076.
- S. K. Manocha and P. Tewari, "Devanagari Handwritten Character Recognition using CNN as Feature Extractor," 2021 International Conference on Smart Generation Computing, Communication and Networking (SMART GENCON), Pune, India, 2021, pp. 1-5, doi: 10.1109/SMARTGENCON51891.2021.9645786.
- MSLB. Subrahmanyam, V. Vijaya Kumar, B. Eswara Reddy, " A Robust Zonal Fractal Dimension Method for the Recognition of Handwritten Telugu Digits," International Journal of Image, Graphics and Signal Processing (IJIGSP), Vol.10, No.9, pp. 42–55, 2018. doi: 10.5815/ijigsp.2018.09.06.
- S. V. Rajashekararadhya and V. P. Ranjan, "Zone-based hybrid feature extraction algorithm for handwritten numeral recognition of four Indian scripts," 2009 IEEE International Conference on Systems, Man and Cybernetics, San Antonio, TX, USA, 2009, pp. 5145–5150, doi: 10.1109/ICSMC.2009.5346007.
- Jyothi, J., Manjusha, K., Anand Kumar, M., & Soman, K. P. (2015). Innovative Feature Sets for Machine Learning based Telugu Character Recognition. Indian Journal of Science and Technology, 8(24), doi: 10.17485/ijst/2015/v8i24/116917.
- R. S. Kunte and S. Samuel, "Script Independent Handwritten Numeral Recognition," IET International Conference on Visual Information Engineering, Bangalore, 2006, pp. 94–98.
Index Terms
Keywords