Deep Learning Approach For Handwritten Gujarati Script Recognition Using Convolutional Neural Network
Abstract
Handwritten character recognition, particularly in languages like Gujarati, poses both opportunities and challenges for technological advancement and cultural preservation. This paper explores the significance of handwritten character recognition in Indian languages, focusing on Gujarati script, which features a complex character set and intricate writing styles. The use of Convolutional Neural Networks (CNNs) in recognizing handwritten characters is discussed, highlighting their ability to extract hierarchical representations from input images and surpass traditional recognition techniques. The architecture and training process of CNNs for handwritten character recognition are detailed, emphasizing their effectiveness in capturing spatial dependencies and structural information in handwritten characters. Literature surveys further demonstrate the growing interest in CNNs for various natural language processing tasks, including sentiment analysis and text classification. The potential for CNN-based techniques to develop unified models for multilingual text processing and improve script identification accuracy is highlighted, suggesting promising directions for future research in artificial intelligence and language technologies.
References
2. Yin, Wenpeng, et al. "Comparative study of CNN and RNN for natural language processing." arXiv preprint arXiv:1702.01923 (2017).
3. Sharma, Amit Kumar, Sandeep Chaurasia, and Devesh Kumar Srivastava. "Sentimental short sentences classification by using CNN deep learning model with fine tuned Word2Vec." Procedia Computer Science 167 (2020): 1139-1147.
4. Widiastuti, Nelly Indriani. "Convolution neural network for text mining and natural language processing." IOP Conference Series: Materials Science and Engineering. Vol. 662. No. 5. IOP Publishing, 2019.
5. Wang, Wei, and Jianxun Gang. "Application of convolutional neural network in natural language processing." 2018 international conference on information Systems and computer aided education (ICISCAE). IEEE, 2018.
6. Maitra, Durjoy Sen, Ujjwal Bhattacharya, and Swapan K. Parui. "CNN based common approach to handwritten character recognition of multiple scripts." 2015 13th International Conference on Document Analysis and Recognition (ICDAR). IEEE, 2015.
7. Cao, Yupeng, et al. "Improving Script Identification by Integrating Text Recognition Information." Aust. J. Intell. Inf. Process. Syst. 16.3 (2019): 67-75.
8. Magare, S. S., Gedam, Y. K., Randhave, D. S., & Deshmukh, R. R. (2014). Character Recognition of Gujarati and Devanagari Script: A Review. International Journal of Engineering Research & Technology (IJERT), 3(01).
9. Pareek, J., Singhania, D., Kumari, R.R. and Purohit, S., 2020. Gujarati handwritten character recognition from text images. Procedia Computer Science, 171, pp.514-523.
10. Patel, C., & Desai, A. (2013). Gujarati handwritten character recognition using hybrid method based on binary tree-classifier and k-nearest neighbour. International Journal of Engineering Research & Technology (IJERT), 2(6), 2337-2345.
11. Naik, V. A., & Desai, A. A. (2017, July). Online handwritten Gujarati character recognition using SVM, MLP, and K-NN. In 2017 8th International Conference on Computing, Communication and Networking Technologies (ICCCNT) (pp. 1-6). IEEE.
12. Macwan, S. J., & Vyas, A. N. (2015, August). Classification of offline Gujarati handwritten characters. In 2015 International Conference on Advances in Computing, Communications and Informatics (ICACCI) (pp. 1535-1541). IEEE.
