Abstract

In many fields, both machine learning and deep learning are being used. One of the most active and difficult areas of research is letter recognition. Recently, deep learning and letter recognition have drawn the attention of many researchers. Letter recognition is a problem that has been worked on in many languages. Letter recognition refers to the process of identifying and distinguishing letters based on their visual features. It involves analyzing patterns of responses and detecting specific components that form different letters. This perceptual activity plays a crucial role in recognizing individual letters and is often studied using rapid visual displays and visual search tasks. Additionally, letter recognition in context explores how letters are processed within words and contributes to our understanding of reading processes. In the process of recognition, pre processing techniques increase image quality by reducing noise and correcting orientation, while convolutional neural networks extract letter features. However, the existing letter recognition system faces many challenges in extracting text from noisy and distorted images or complex layouts, and extraction is mostly limited to numbers and the English alphabet. So, several studies using deep learning models have been conducted to achieve better accuracy. The proposed technique achieves 96.2% accuracy in recognizing letters from input images.

Keywords

  • Deep Learning
  • Convolutional Neural Network (CNN)
  • Letter Recognition
  • Noisy images

References

  1. 1. Srivastava, Ankur, et al. "Handwritten character recognition using machine learning and deep learning." Emerging Trends in IoT and Computing Technologies. Routledge, 2023. 475-482.
  2. 2. Bhatt, Preeti P., and Isha Patel. "Optical character recognition using deep learning-a technical review." National Journal of System and Information Technology 11.1 (2018): 55.
  3. 3. Saqib, Nazmus, et al. "Convolutional-neural-network-based handwritten character recognition: an approach with massive multisource data." Algorithms 15.4 (2022): 129.
  4. 4. Younis, Khaled S. "Arabic hand-written character recognition based on deep convolutional neural networks." Jordanian Journal of Computers and Information Technology 3.3 (2017).
  5. 5. Balaha, Hossam Magdy, et al. "A new Arabic handwritten character recognition deep learning system (AHCR-DLS)." Neural Computing and Applications 33 (2021): 6325-6367.
  6. 6. Munappy, Aiswarya Raj, et al. "Data management for production quality deep learning models: Challenges and solutions." Journal of Systems and Software 191 (2022): 111359.
  7. 7. Liman, Muhamad Arief, Antonio Josef, and Gede Putra Kusuma. "Handwritten character recognition using deep learning algorithm with machine learning classifier." JOIV: International Journal on Informatics Visualization 8.1 (2024): 150-157.
  8. 8. Reutzel, Pamela, Kathleen AJ Mohr, and Cindy D. Jones. "Exploring the relationship between letter recognition and handwriting in early literacy development." Journal of Early Childhood Literacy 19.3 (2019): 349-374.
  9. 9. Neisser, Ulric, and Paul Weene. "A note on human recognition of hand-printed characters." Information and Control 3.2 (1960): 191-196.
  10. 10. Naus, Mary J., and Robert J. Shillman. "Why a Y is not a V: a new look at the distinctive features of letters." Journal of Experimental Psychology: Human Perception and Performance 2.3 (1976): 394.
  11. 11. Neisser, Ulric, and Henry K. Beller. "Searching through word lists." British Journal of Psychology 56.4 (1965): 349-358.
  12. 12. Pelli, Denis G., Bart Farell, and Deborah C. Moore. "The remarkable inefficiency of word recognition." Nature 423.6941 (2003): 752-756.
  13. 13. Gibson, Eleanor P., et al. "A developmental study of the discrimination of letter-like forms." Journal of comparative and physiological psychology 55.6 (1962): 897.
  14. 14. Pelli, Denis G., et al. "Identifying letters." Vision Research 46.28 (2006): 4646-4674.
  15. 15. Sethi, Rohan, and Ila Kaushik. "Handwritten digit recognition using machine learning." 2020 IEEE 9th International Conference on Communication Systems and Network Technologies (CSNT). IEEE, 2020.
  16. 16. Pashine, Samay, Ritik Dixit, and Rishika Kushwah. "Handwritten digit recognition using machine and deep learning algorithms." arXiv preprint arXiv:2106.12614 (2021).
  17. 17. Rai, Ankur, Siddharth Singh, and Sumit Kushwaha. "Optimized Handwritten Digit Recognition: A Convolutional Neural Network Approach." 2024 International Conference on Communication, Control, and Intelligent Systems (CCIS). IEEE, 2024.
  18. 18. Shetty, Ashish, and Sanjeev Sharma. "Ensemble deep learning model for optical character recognition." Multimedia Tools and Applications 83.4 (2024): 11411-11431.
  19. 19. Rakesh, S., et al. "Handwritten text recognition using deep learning techniques: A survey." MATEC Web of Conferences. Vol. 392. EDP Sciences, 2024.
  20. 20. Graves, Alex, et al. "A novel connectionist system for unconstrained handwriting recognition." IEEE transactions on pattern analysis and machine intelligence 31.5 (2008): 855-868.
  21. 21. Nikitha, A., J. Geetha, and D. S. JayaLakshmi. "Handwritten text recognition using deep learning." 2020 International Conference on Recent Trends on Electronics, Information, Communication & Technology (RTEICT). IEEE, 2020.
  22. 22. Graves, Alex, and Jürgen Schmidhuber. "Offline handwriting recognition with multidimensional recurrent neural networks." Advances in neural information processing systems 21 (2008).
  23. 23. Liwicki, Marcus, et al. "A novel approach to on-line handwriting recognition based on bidirectional long short-term memory networks." Proceedings of the 9th International Conference on Document Analysis and Recognition, ICDAR 2007. 2007.
  24. 24. LeCun, Yann, et al. "Gradient-based learning applied to document recognition." Proceedings of the IEEE 86.11 (1998): 2278-2324.
  25. 25. Arora, Sandhya, et al. "Devanagari Character Recognition: A Comprehensive Literature Review." IEEE Access (2024).
  26. 26. Madhvanath, Sriganesh, and Venu Govindaraju. "The role of holistic paradigms in handwritten word recognition." IEEE transactions on pattern analysis and machine intelligence 23.2 (2001): 149-164.
  27. 27. Fanany, Mohamad Ivan. "Handwriting recognition on form document using convolutional neural network and support vector machines (CNN-SVM)." 2017 5th international conference on information and communication technology (ICoIC7). IEEE, 2017.
  28. 28. Balaha, Hossam Magdy, et al. "A new Arabic handwritten character recognition deep learning system (AHCR-DLS)." Neural Computing and Applications 33 (2021): 6325-6367.
  29. 29. Younis, Khaled S. "Arabic hand-written character recognition based on deep convolutional neural networks." Jordanian Journal of Computers and Information Technology 3.3 (2017).
  30. 30. Bora, Mayur Bhargab, et al. "Handwritten character recognition from images using CNN-ECOC." Procedia Computer Science 167 (2020): 2403-2409.
  31. 31. Lahreche, Afafe, and Abdelhakim Cheriet. "Deep learning for Arabic letters recognition." (2019).