Syllable-Based Detection and Recognition of Braille Cells Using Faster R-CNN
DOI:
https://doi.org/10.24269/mtkind.v17i1.13429Keywords:
Braille Recognition, Computer Vision, Faster R-CNN, Image Processing, Syllable-Based DetectionAbstract
Braille digitization plays an important role in improving access to written information for visually impaired individuals. However, automatic recognition of Braille text remains challenging due to the small size, dense spatial arrangement, and subtle variations of Braille dot patterns. This study proposes a syllable-based Braille cell detection and recognition framework using a customized Faster R-CNN architecture. The system employs a Region Proposal Network (RPN) to localize individual Braille cells, followed by an AlexNet-based classifica-tion network to recognize syllable-level patterns. The proposed method is evaluated on a syllable-level Braille dataset covering 50 syllable classes, and a two-stage training strategy is adopted to improve bounding box localization accuracy. Experimental results show stable training convergence and consistent classification performance across syllable categories. Confusion matrix analysis indicates that most misclassifications occur among syllables with visually similar dot configurations. Despite sensitivity to variations in physical Braille quality, the proposed framework indicates potential applicability in accessibility-oriented Braille digitization systems
Downloads
References
[1] T. Kausar, S. Manzoor, A. Kausar, Y. Lu, M. Wasif, and M. A. Ashraf, “Deep Learning Strategy for Braille Character Recognition,” IEEE Access, vol. 9, pp. 169357–169371, 2021, doi: 10.1109/ACCESS.2021.3138240.
[2] L. Lu, D. Wu, J. Xiong, Z. Liang, and F. Huang, “Anchor-Free Braille Character Detection Based on Edge Feature in Natural Scene Images,” Comput. Intell. Neurosci., vol. 2022, pp. 1–11, Aug. 2022, doi: 10.1155/2022/7201775.
[3] M. Kim, H. Shin, and M. Jekal, “Braille glove design toward interdisciplinary approach for visually impaired people: developing independent sensing design with MXene and embroidery,” Fashion and Textiles, vol. 11, no. 1, p. 18, Apr. 2024, doi: 10.1186/s40691-024-00383-x.
[4] M. A. D. Widyadara, A. N. Handayani, H. W. Herwanto, and T. Yu, “A Generalized Deep Learning Approach for Multi Braille Character (MBC) Recognition,” Buletin Ilmiah Sarjana Teknik Elektro, vol. 7, no. 3, pp. 434–449, Aug. 2025, doi: 10.12928/biste.v7i3.13891.
[5] Y.-P. Chen, Y. Li, G. Wang, and Q. Xu, “A Multi-strategy Region Proposal Network,” Expert Syst. Appl., vol. 113, pp. 1–17, Dec. 2018, doi: 10.1016/j.eswa.2018.06.043.
[6] K. A. S. K. and B. G., “Faster region based convolution neural network with context iterative refinement for object detection,” Measurement: Sensors, vol. 31, p. 101025, Feb. 2024, doi: 10.1016/j.measen.2024.101025.
[7] S. Ren, K. He, R. Girshick, and J. Sun, “Faster R-CNN: Towards Real-Time Object Detection with Region Proposal Networks,” IEEE Trans. Pattern Anal. Mach. Intell., vol. 39, no. 6, pp. 1137–1149, Jun. 2017, doi: 10.1109/TPAMI.2016.2577031.
[8] M. Jamali, P. Davidsson, R. Khoshkangini, M. G. Ljungqvist, and R.-C. Mihailescu, “Context in object detection: a systematic literature review,” Artif. Intell. Rev., vol. 58, no. 6, p. 175, Mar. 2025, doi: 10.1007/s10462-025-11186-x.
[9] H. Yao et al., “Development and optimization of object detection technology in pavement engineering: A literature review,” Journal of Road Engineering, vol. 4, no. 2, pp. 163–188, Jun. 2024, doi: 10.1016/j.jreng.2024.01.006.
[10] S. M. Sait, A. El-Maleh, M. Altakrouri, and A. Shawahna, “Optimization of FPGA-based CNN Accelerators Using Metaheuristics,” Sep. 2022, doi: 10.1007/s11227-022-04787-8.
[11] X. Zhao, L. Wang, Y. Zhang, X. Han, M. Deveci, and M. Parmar, “A review of convolutional neural networks in computer vision,” Artif. Intell. Rev., vol. 57, no. 4, p. 99, Mar. 2024, doi: 10.1007/s10462-024-10721-6.
[12] J. Liu, T. Xing, and X. Wang, “Optimizing AlexNet for accurate tree species classification via multi-branch architecture and mixed-domain attention,” Complex & Intelligent Systems, vol. 11, no. 6, p. 244, Jun. 2025, doi: 10.1007/s40747-025-01836-6.
[13] Q. Chaudhary et al., “SAlexNet: Superimposed AlexNet using residual attention mechanism for accurate and efficient automatic primary brain tumor detection and classification,” Results in Engineering, vol. 25, p. 104025, Mar. 2025, doi: 10.1016/j.rineng.2025.104025.
[14] M. Z. Alom et al., “The History Began from AlexNet: A Comprehensive Survey on Deep Learning Approaches,” Sep. 2018.
[15] E. Edozie, A. N. Shuaibu, U. K. John, and B. O. Sadiq, “Comprehensive review of recent developments in visual object detection based on deep learning,” Artif. Intell. Rev., vol. 58, no. 9, p. 277, Jun. 2025, doi: 10.1007/s10462-025-11284-w.
[16] G. Cui and L. Zhang, “Improved faster region convolutional neural network algorithm for UAV target detection in complex environment,” Results in Engineering, vol. 23, p. 102487, Sep. 2024, doi: 10.1016/j.rineng.2024.102487.
[17] R. Girshick, “Fast R-CNN,” Sep. 2015.
[18] Z. Wang and J. Wang, “Personalized icon design model based on improved Faster-RCNN,” Systems and Soft Computing, vol. 7, p. 200193, Dec. 2025, doi: 10.1016/j.sasc.2025.200193.
[19] R. Girshick, “Fast R-CNN,” Apr. 2015, [Online]. Available: http://arxiv.org/abs/1504.08083
[20] A. Dutta, “Object Detection Technique Using a Fast R-CNN Based Technique,” Int. J. Res. Appl. Sci. Eng. Technol., vol. 10, no. 6, pp. 1683–1687, Jun. 2022, doi: 10.22214/ijraset.2022.44112.
[21] L. Aziz, Md. S. Bin Haji Salam, U. U. Sheikh, and S. Ayub, “Exploring Deep Learning-Based Architecture, Strategies, Applications and Current Trends in Generic Object Detection: A Comprehensive Review,” IEEE Access, vol. 8, pp. 170461–170495, 2020, doi: 10.1109/ACCESS.2020.3021508.
[22] J. R. R. Uijlings, K. E. A. van de Sande, T. Gevers, and A. W. M. Smeulders, “Selective Search for Object Recognition,” Int. J. Comput. Vis., vol. 104, no. 2, pp. 154–171, Sep. 2013, doi: 10.1007/s11263-013-0620-5.
[23] M. I. Rakib et al., “TWIG: Two-Step Image Generation using Segmentation Masks in Diffusion Models,” May 2025, doi: 10.1007/978-3-031-91054-8_4.
[24] Q. Chen, M. Li, Z. Lai, J. Zhu, and L. Guan, “A Multi-Scale Target Detection Method Using an Improved Faster Region Convolutional Neural Network Based on Enhanced Backbone and Optimized Mechanisms,” J. Imaging, vol. 10, no. 8, p. 197, Aug. 2024, doi: 10.3390/jimaging10080197.
[25] P. Steno, A. Alsadoon, P. W. C. Prasad, T. Al-Dala’in, and O. H. Alsadoon, “A novel enhanced region proposal network and modified loss function: threat object detection in secure screening using deep learning,” J. Supercomput., vol. 77, no. 4, pp. 3840–3869, Apr. 2021, doi: 10.1007/s11227-020-03418-4.
[26] M. Li, Y. Hu, H. Chen, and X. Li, “Feature-Enhanced Faster R-CNN for Small Object Detection in Complex Scenes,” IEEE Access, vol. 13, pp. 174297–174315, 2025, doi: 10.1109/ACCESS.2025.3618019.
[27] S. Pha, H. Zheng, and Y. Sun, “Research on Detection Method of Insulator Image Based on Improved Faster R-CNN,” J. Phys. Conf. Ser., vol. 2213, no. 1, p. 012036, Mar. 2022, doi: 10.1088/1742-6596/2213/1/012036.
[28] Z. Jia, Z. Shi, Z. Quan, and M. Shunqi, “Fabric defect detection based on transfer learning and improved Faster R-CNN,” J. Eng. Fiber. Fabr., vol. 17, Jan. 2022, doi: 10.1177/15589250221086647.
[29] S. M. Abbas and Dr. S. N. Singh, “Region-based Object Detection and Classification using Faster R-CNN,” in 2018 4th International Conference on Computational Intelligence & Communication Technology (CICT), IEEE, Feb. 2018, pp. 1–6. doi: 10.1109/CIACT.2018.8480413.
[30] S. Ren, K. He, R. Girshick, and J. Sun, “Faster R-CNN: Towards Real-Time Object Detection with Region Proposal Networks,” Jun. 2015, [Online]. Available: http://arxiv.org/abs/1506.01497
[31] H. M. Al-Barhamtoshy, K. M. Jambi, M. A. Rashwan, and S. M. Abdou, “An Arabic Manuscript Regions Detection, Recognition and Its Applications for OCRing,” ACM Transactions on Asian and Low-Resource Language Information Processing, vol. 22, no. 1, pp. 1–28, Jan. 2023, doi: 10.1145/3532609.
[32] Y. Sun, C. Sun, D. Wang, Y. He, and H. Lu, “ROI Pooled Correlation Filters for Visual Tracking,” in 2019 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR), IEEE, Jun. 2019, pp. 5776–5784. doi: 10.1109/CVPR.2019.00593.
[33] Z. B. Alawi, “A Comparative Survey of PyTorch vs TensorFlow for Deep Learning: Usability, Performance, and Deployment Trade-offs,” Aug. 2025.
[34] V. Venkatesh Murthy and M. Hanumanthappa, “VGG16 CNN based Braille Cell classifier model for Translation of Braille to Text.”
[35] S. Tiendee, C. Lerdsudwichai, S. Thainimit, and C. Sinthanayothin, “The Method of Braille Embossed Dots Segmentation for Braille Document Images Produced on Reusable Paper,” International Journal of Advanced Computer Science and Applications, vol. 13, no. 2, 2022, doi: 10.14569/IJACSA.2022.0130220.
Downloads
Published
Issue
Section
License
Copyright (c) 2026 Made Ayu Dusea Widyadara; Risky Aswi Ramadhani
This work is licensed under a Creative Commons Attribution-ShareAlike 4.0 International License.
MULTITEK is licensed under a Creative Commons Attribution-NonCommercial-ShareAlike 4.0 International License
