Real-Time Text Extraction and Classification from Bilingual Road Signboards Using OCR Engines.
Keywords:
EasyOCR, Image processing, MSER, Performance metric, Road signboards, Tesseract.Abstract
The objective of the project is to develop a system that employs image processing methods to retrieve text from multilingual roadway directional signs. Multilingual signboards with language overlap, inconsistent fonts, and noisy real-time images complicate automated text extraction in various regions (English, Hindi, Kannada, etc.). This project includes efficient image preprocessing methods to improve the clarity of live images. Two OCR engines EasyOCR and Tesseract—are employed to extract the entire text content, subsequently categorized into English and non-English groups. To enhance the evaluation of the system, a specialized performance metric module has been established. This module examines the speed and reliability of both OCR engines through processing time. Visual depictions like bar charts and line graphs have been incorporated to assess the engines' performance and determine the quicker and more dependable choice. The incorporation of this performance analysis offers a more thorough insight into the system’s functioning and practical relevance.
Downloads
References
S. Das et al., “Bangla signboard understanding using deep learning,” IEEE Access, vol. 11, pp. 23456–23467, 2023.
T. Saha and M. Sharma, “Machine learning-based multilingual signboard translation,” International Journal of Computer Applications, vol. 184, no. 12, pp. 45–51, 2022.
A. Mohamed et al., “Real-time OCR for road signs using dashcam input,” Malaysian Journal of Computer Science, vol. 35, no. 1, pp. 50–62, 2022.
A. K. Bhunia et al., “E2E-MLT: An unconstrained end-to-end method for multi-language scene text,” Pattern Recognition, vol. 122, 108255, 2021.M. Naderi et al., “End-to-end Arabic-English bilingual scene text recognition,” Multimedia Tools and Applications, vol. 80, pp. 34375–34397, 2021.
B. Shi et al., “Script identification using deep attention-based CNN,” ICDAR, pp. 679–684, 2020.
Y. Yin et al., “ICDAR 2019 competition on Chinese text recognition in the wild,” ICDAR Competition Report, 2019.
R. Sharma et al., “Unified model for multilingual scene text detection and recognition,” IEEE Transactions on Image Processing, vol. 26, no. 12, pp. 5795–5808, 2017.
P. Krishnan et al., “Script identification in document images using deep CNN,” ICDAR, pp. 1051–1056, 2017.
A. Roy et al., “Signboard detection for smart vehicles using Tesseract OCR,” IET Intelligent Transport Systems, vol. 10, no. 6, pp. 391–397, 2016.
M. Jaderberg et al., “Reading text in the wild with convolutional neural networks,” International Journal of Computer Vision, vol. 116, pp. 1–20, 2016.
Y. Chen et al., “Directional Road sign detection using mobile mapping and OCR,” Computer-Aided Civil and Infrastructure Engineering, vol. 29, no. 7, pp. 501–517, 2014.
A. Patil et al., “Text detection in signboards using morphological operations,” IJSER, vol. 4, no. 5, pp. 123–128, 2013.
P. Singh and A. Kumar, “Hindi OCR using neural networks,” Journal of Emerging Technologies, vol. 8, no. 2, pp. 45–50, 2013.
C. Leung et al., “Real-time text extraction from videos for road signs,” IEEE Transactions on Image Processing, vol. 15, no. 11, pp. 3277–3288, 2006.
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.
