A Unified Tesseract-Based Text-To-Braille Conversion System For Visually Impaired People
Main Article Content
Abstract
Background: Braille has been the most common means of reading and learning among visually challenged people for many years. Despite the importance of developing Braille-assistive technologies for blind people, the there exists a clear gap in the current Optical Character Recognition (OCR)-based conversion of Text-to-Braille documents and accuracy of translator systems. While the Long Short-Term Memory (LSTM)-based Tesseract OCR engine has been a popular tool in language processing, its ability in capturing mixed alphabet and mathematical symbols from a scanned document has not been explored earlier.
Methods: This work presents a novel Tesseract LSTM-based Text-to-Brielle conversion framework that converts English alphabets, mathematical symbols, punctuations, sentences, paragraphs in a scanned image to Grade-1 and Grade-2 Braille codes for visually challenged people. Using SimpleCV image pre-processing algorithms the image filtering, detection of edges and segmentation of mixed characters were accomplished, and the performance of the Tesseract LSTM model was examined by testing its accuracy in converting mixed fonts or symbol of various sizes, and converting whole document image to Braille codes.
Results: The results revealed that the efficacy of Tesseract OCR engine for Brielle conversion was better for large size font and symbol, showing 86% accuracy for fonts greater than 16 point size with an average accuracy of 88%. It was observed that the economic conversion of Text-to-Braille using open source LSTM network provide a powerful tool for fast and language-free translation of texts to both Grade 1 and 2 Braille.
Conclusion: The Tesseract OCR engine provide an efficient, cost-effective way of converting mixed text or document images to Braille codes, and can be extended easily to build real-time multilingual text-to-Braille translators for visually impaired.
Article Details
References
F. S. Apu, F. I. Joyti, M. A. U. Anik, M. W. U. Zobayer, A. K. Dey, and S. Sakhawat, “Text and Voice to Braille Translator for Blind People,” in 2021 International Conference on Automation, Control and Mechatronics for Industry 4.0 (ACMI), Jul. 2021, pp. 1–6. doi: 10.1109/ACMI53878.2021.9528283.
G. G. Devi and G. Sathyanarayanan, “Braille Document Recognition in Southern Indian Languages– A Review,” in 2018 Fourth International Conference on Advances in Electrical, Electronics, Information, Communication and Bio-Informatics (AEEICB), Feb. 2018, pp. 1–4. doi: 10.1109/AEEICB.2018.8480950.
F. S. Apu, F. I. Joyti, M. A. U. Anik, M. W. U. Zobayer, A. K. Dey, and S. Sakhawat, “Text and Voice to Braille Translator for Blind People,” in 2021 International Conference on Automation, Control and Mechatronics for Industry 4.0 (ACMI), Jul. 2021, pp. 1–6. doi: 10.1109/ACMI53878.2021.9528283.
P. Blenkhorn, “A system for converting braille into print,” IEEE Transactions on Rehabilitation Engineering, vol. 3, no. 2, pp. 215–221, Jun. 1995, doi: 10.1109/86.392366.
P. Blenkhorn and G. Evans, “Automated Braille production from word-processed documents,” IEEE Transactions on Neural Systems and Rehabilitation Engineering, vol. 9, no. 1, pp. 81–85, Mar. 2001, doi: 10.1109/7333.918280.
S. Shetty, S. Shetty, S. Hegde, and K. Pandit, “Transliteration of text input from Kannada to Braille and vice versa,” in 2019 IEEE International Conference on Distributed Computing, VLSI, Electrical Circuits and Robotics (DISCOVER), Aug. 2019, pp. 1–4. doi: 10.1109/DISCOVER47552.2019.9007909.
M. V.Umarani and R. P.Sheddi, “A Review of Kannada Text to Braille Conversion,” 2018.https://www.semanticscholar.org/paper/A-Review-of-Kannada-Text-to-Braille-Conversion-V.Umarani-P.Sheddi/dfeabc9d90f38821b11017bda01cd9266f3e84d8
“A Model for Translation of Text from Indian Languages to Bharti Braille Characters | IEEE Conference Publication | IEEE Xplore.” https://ieeexplore.ieee.org/document/10112021
P. Vashist et al., “Blindness and visual impairment and their causes in India: Results of a nationally representative survey,” PLoS One, vol. 17, no. 7, p. e0271736, 2022, doi: 10.1371/journal.pone.0271736.
C. N. R. Kumar and S. Srinath, “A Novel and Efficient Algorithm to Recognize Any Universally Accepted Braille Characters: A Case with Kannada Language,” 2014 Fifth International Conference on Signal and Image Processing, pp. 292–296, Jan. 2014, doi: 10.1109/ICSIP.2014.52.
T. D. S. H. Perera and W. K. I. L. Wanniarachchi, “Optical Braille Translator for Sinhala Braille System: Paper Communication Tool Between Vision Impaired and Sighted Persons,” Nov. 26, 2018, Rochester, NY: 3290477. Accessed: Jul. 13, 2024. [Online]. Available: https://papers.ssrn.com/abstract=3290477
L. Hakobyan, J. Lumsden, D. O’Sullivan, and H. Bartlett, “Mobile assistive technologies for the visually impaired,” Surv Ophthalmol, vol. 58, no. 6, pp. 513–528, 2013, doi: 10.1016/j.survophthal.2012.10.004.
D. T. V. Pawluk, R. J. Adams, and R. Kitada, “Designing Haptic Assistive Technology for Individuals Who Are Blind or Visually Impaired,” IEEE Trans Haptics, vol. 8, no. 3, pp. 258–278, 2015, doi: 10.1109/TOH.2015.2471300.
“Tesseract: an Open-Source Optical Character Recognition Engine | Linux Journal.” 2016. https://www.linuxjournal.com/article/96