Analysis of Recent Deep Learning Techniques for Arabic Handwritten-Text OCR and Post-OCR Correction
Source abstract
Abstract
Arabic handwritten-text recognition applies an OCR technique and then a text-correction technique to extract the text within an image correctly. Deep learning is a current paradigm utilized in OCR techniques. However, no study investigated or critically analyzed recent deep-learning techniques used for Arabic handwritten OCR and text correction during the period of 2020–2023. This analysis fills this noticeable gap in the literature, uncovering recent developments and their limitations for researchers, practitioners, and interested readers. The results reveal that CNN-LSTM-CTC is the most suitable architecture among Transformer and GANs for OCR because it is less complex and can hold long textual dependencies. For OCR text correction, applying DL models to generated errors in datasets improved accuracy in many works. In conclusion, Arabic OCR has the potential to further apply several text-embedding models to correct the resultant text from the OCR, and there is a significant gap in studies investigating this problem. In addition, there is a need for more high-quality and domain-specific OCR Arabic handwritten datasets. Moreover, we recommend the practical development of a space for future trends in Arabic OCR applications, derived from current limitations in Arabic OCR works and from applications in other languages; this will involve a plethora of possibilities that have not been effectively researched at the time of writing.
Source: crossref · Quality: baseline
Research context
OpenAlex topics: Handwritten Text Recognition Techniques · Image Processing and 3D Reconstruction · Infrastructure Maintenance and Monitoring
SciVX discovery tags are terminology-based. OpenAlex topics are source-provided classifications.
Affiliations
Department of Computer Science, Islamic University, Madinah 42351, Saudi Arabia
Funding
Deputyship of Research and Innovation, Ministry of Education, Saudi Arabia · Islamic University of Madinah
References (83)
Show source-provided references
1. Raj, R., and Kos, A. (2022, January 25–27). A Comprehensive Study of Optical Character Recognition. Proceedings of the 29th International Conference on Mixed Design of Integrated Circuits and System (MIXDES), Lodz, Poland.10.23919/MIXDES55591.2022.9837974 ↗
2. Evaluating and mitigating the impact of OCR errors on information retrieval10.1007/s00799-023-00345-6 ↗
3. Xu, Q., Wang, L., Liu, H., and Liu, N. (2022, January 21–23). LayoutLM-Critic: Multimodal Language Model for Text Error Correction of Optical Character Recognition. Proceedings of the Artificial Intelligence and Robotics 7th International Symposium, Shanghai, China.10.1007/978-981-19-7943-9_11 ↗
4. Hajiali, M., Cacho, J.R.F., and Taghva, K. (2022). Intelligent Computing, Proceedings of the 2021 Computing Conference, Online, 15–16 July 2021, Springer.
5. Ravi, S., Chauhan, S., Yadlapallii, S.H., Jagruth, K., and Manikandan, V.M. (2022). Lecture Notes in Networks and Systems, Proceedings of the 13th International Conference on Soft Computing and Pattern Recognition (SoCPaR 2021), Online, 15–17 December 2021, Springer.
6. Artificial Intelligence image recognition using self-regulation learning strategies: Effects on vocabulary acquisition, learning anxiety, and learning behaviours of English language learners10.1080/10494820.2023.2165508 ↗
7. Bhattamisra, S.K., Banerjee, P., Gupta, P., Mayuren, J., Patra, S., and Candasamy, M. (2023). Artificial Intelligence in Pharmaceutical and Healthcare Research. Big Data Cogn. Comput., 7.10.3390/bdcc7010010 ↗
8. Ranjan, S., Sanket, S., Singh, S., Tyagi, S., Kaur, M., Rakesh, N., and Nand, P. (2022, January 23–25). OCR based Automated Number Plate Text Detection and Extraction. Proceedings of the 9th International Conference on Computing for Sustainable Global Development (INDIACom), New Delhi, India.10.23919/INDIACom54597.2022.9763248 ↗
9. BLPnet: A new DNN model and Bengali OCR engine for Automatic Licence Plate Recognition10.1016/j.array.2022.100244 ↗
10. Azadbakht, A., Kheradpisheh, S.R., and Farahani, H. (2022, January 17–18). MultiPath ViT OCR: A Lightweight Visual Transformer-based License Plate Optical Character Recognition. Proceedings of the 12th International Conference on Computer and Knowledge Engineering (ICCKE), Mashhad, Iran.10.1109/ICCKE57176.2022.9960026 ↗
11. Bi, S., Wang, C., Zhang, J., Huang, W., Wu, B., Gong, Y., and Ni, W. (2022). A Survey on Artificial Intelligence Aided Internet-of-Things Technologies in Emerging Smart Libraries. Sensors, 22.10.3390/s22082991 ↗
12. A novel offline handwritten text recognition technique to convert ruled-line text into digital text through deep neural networks10.1007/s11042-022-12097-7 ↗
13. Singh, T.P., Gupta, S., and Garg, M. (2022, January 13). A Review on Online and Offline Handwritten Gurmukhi Character Recognition. Proceedings of the 10th International Conference on Reliability, Infocom Technologies and Optimization (Trends and Future Directions) (ICRITO), Noida, India.10.1109/ICRITO56286.2022.9964657 ↗
14. Tan, Y.F., Connie, T., Goh, M.K.O., and Teoh, A.B.J. (2022). A Pipeline Approach to Context-Aware Handwritten Text Recognition. Appl. Sci., 12.10.3390/app12041870 ↗
15. Ott, F., Rügamer, D., Heublein, L., Bischl, B., and Mutschler, C. (2023). Representation Learning for Tablet and Paper Domain Adaptation in Favor of Online Handwriting Recognition. arXiv.10.1007/978-3-031-37660-3_26 ↗
16. Advances in online handwritten recognition in the last decades10.1016/j.cosrev.2022.100515 ↗
17. Statista (2023, February 06). The Most Spoken Languages Worldwide 2022. Available online: https://www.statista.com/statistics/266808/the-most-spoken-languages-worldwide/.
18. PRo-Pat: Probabilistic Root–Pattern Bi-gram data language model for Arabic based morphological analysis and distribution10.1016/j.dib.2022.108875 ↗
19. A Review of Arabic Optical Character Recognition Techniques & Performance
20. Quranic Optical Text Recognition Using Deep Learning Models10.1109/ACCESS.2021.3064019 ↗
21. Arabic Handwritten Recognition Using Deep Learning: A Survey10.1007/s13369-021-06363-3 ↗
22. Moudgil, A., Singh, S., and Gautam, V. (2022). Machine Learning for Edge Computing, CRC Press.
23. Avyodri, R., Lukas, S., and Tjahyadi, H. (2022, January 23). Optical Character Recognition (OCR) for Text Recognition and its Post-Processing Method: A Literature Review. Proceedings of the 1st International Conference on Technology Innovation and Its Applications (ICTIIA), Tangerang, Indonesia.10.1109/ICTIIA54654.2022.9935961 ↗
24. A Comprehensive Survey of Image Augmentation Techniques for Deep Learning10.1016/j.patcog.2023.109347 ↗
25. The Significance of Machine Learning and Deep Learning Techniques in Cybersecurity: A Comprehensive Review
26. Has the Future Started? The Current Growth of Artificial Intelligence, Machine Learning, and Deep Learning
27. Deep Learning (CNN) and Transfer Learning: A Review10.1088/1742-6596/2273/1/012029 ↗
28. Setyanto, A., Laksito, A., Alarfaj, F., Alreshoodi, M., Oyong, I., Hayaty, M., Alomair, A., Almusallam, N., and Kurniasari, L. (2022). Arabic Language Opinion Mining Based on Long Short-Term Memory (LSTM). Appl. Sci., 12.10.3390/app12094140 ↗
29. Atenco, J.C., Moreno, J.C., and Ramirez, J.M. (2023). Audiovisual Biometric Network with Deep Feature Fusion for Identification and Text Prompted Verification. Algorithms, 16.10.3390/a16020066 ↗
30. Zhou, C., Li, Q., Li, C., Yu, J., Liu, Y., Wang, G., Zhang, K., Ji, C., Yan, Q., and He, L. (2023). A Comprehensive Survey on Pretrained Foundation Models: A History from BERT to ChatGPT. arXiv.
31. Exploring the Limits of Transfer Learning with a Unified Text-to-Text Transformer
32. Brown, T., Mann, B., Ryder, N., Subbiah, M., Kaplan, J.D., Dhariwal, P., Neelakantan, A., Shyam, P., Sastry, G., and Askell, A. (2020, January 6–12). Language Models are Few-Shot Learners. Proceedings of the Advances in Neural Information Processing Systems, New Orleans, LA, USA. Available online: https://papers.nips.cc/paper/2020/hash/1457c0d6bfcb4967418bfb8ac142f64a-Abstract.html.
33. Zhuo, T.Y., Huang, Y., Chen, C., and Xing, Z. (2023). Exploring AI Ethics of ChatGPT: A Diagnostic Analysis. arXiv.
34. (2023, February 06). CER-a Hugging Face Space by Evaluate-Metric. Available online: https://huggingface.co/spaces/evaluate-metric/cer.
35. (2023, February 06). WER-a Hugging Face Space by Evaluate-Metric. Available online: https://huggingface.co/spaces/evaluate-metric/wer.
36. An Enhanced Offline Printed Arabic OCR Model Based on Bio-Inspired Fuzzy Classifier10.1109/ACCESS.2020.3004286 ↗
37. Emon, I.H., Iqbal, K.N., Mehedi, H.K., Mahbub, M.J.A., and Rasel, A.A. (2023). Emerging Technologies in Computing, Springer.
38. Characters as graphs: Interpretable handwritten Chinese character recognition via Pyramid Graph Transformer10.1016/j.patcog.2023.109317 ↗
39. Feature extraction and classification techniques for handwritten Devanagari text recognition: A survey10.1007/s11042-022-13318-9 ↗
40. Hijji, M., Iqbal, R., Pandey, A.K., Doctor, F., Karyotis, C., Rajeh, W., Alshehri, A., and Aradah, F. (2023). 6G Connected Vehicle Framework to Support Intelligent Road Maintenance Using Deep Learning Data Fusion. IEEE Trans. Intell. Transp. Syst., 1–10.10.1109/TITS.2023.3235151 ↗
41. Deep learning for power quality10.1016/j.epsr.2022.108887 ↗
42. Applications of generative adversarial networks in neuroimaging and clinical neuroscience10.1016/j.neuroimage.2023.119898 ↗
43. Review of deep learning: Concepts, CNN architectures, challenges, applications, future directions10.1186/s40537-021-00444-8 ↗
44. End-to-End Sequence Labeling via Convolutional Recurrent Neural Network with a Connectionist Temporal Classification Layer10.2991/ijcis.d.200316.001 ↗
45. Vaswani, A., Shazeer, N., Parmar, N., Uszkoreit, J., Jones, L., Gomez, A.N., Kaiser, Ł., and Polosukhin, I. (2017, January 4–9). Attention is All you Need. Proceedings of the Advances in Neural Information Processing Systems, Long Beach, CA, USA. Available online: https://proceedings.neurips.cc/paper/2017/hash/3f5ee243547dee91fbd053c1c4a845aa-Abstract.html.
46. Sajun, A.R., and Zualkernan, I. (2022). Survey on Implementations of Generative Adversarial Networks for Semi-Supervised Learning. Appl. Sci., 12.10.3390/app12031718 ↗
47. Noubigh, Z., Mezghani, A., and Kherallah, M. (2020, January 28–30). Transfer Learning to improve Arabic handwriting text Recognition. Proceedings of the 21st International Arab Conference on Information Technology (ACIT), Giza, Egypt.10.1109/ACIT50332.2020.9300105 ↗
48. Boualam, M., Elfakir, Y., Khaissidi, G., and Mrabti, M. (2022). WITS 2020, Proceeding of the 6th International Conference on Wireless Technologies, Embedded, and Intelligent Systems, Fez, Morocco, 14–16 October 2022, Springer.
49. Shtaiwi, R.E., Abandah, G.A., and Sawalhah, S.A. (2022, January 21–23). End-to-End Machine Learning Solution for Recognizing Handwritten Arabic Documents. Proceedings of the 13th International Conference on Information and Communication Systems (ICICS), Irbid, Jordan.10.1109/ICICS55353.2022.9811155 ↗
50. Alzrrog, N., Bousquet, J.-F., and El-Feghi, I. (2022, January 18–20). Deep Learning Application for Handwritten Arabic Word Recognition. Proceedings of the IEEE Canadian Conference on Electrical and Computer Engineering (CCECE), Halifax, NS, Canada.10.1109/CCECE49351.2022.9918375 ↗
51. Arabic (Indian) digit handwritten recognition using recurrent transfer deep architecture10.1007/s00500-020-05368-8 ↗
52. A Hybrid Deep Learning Model for Arabic Text Recognition
53. A deep learning model for Ottoman OCR10.1002/cpe.6937 ↗
54. Khaled, M.M., Alzebdeh, A., Lataifeh, M., Lulu, L., and Elnagar, A.M. (2023). A Hybrid Deep Learning Approach for Arabic Handwritten Recognition: Exploring the Complexities of the Arabic Language, Rochester.10.2139/ssrn.4399243 ↗
55. Mostafa, A., Mohamed, O., Ashraf, A., Elbehery, A., Jamal, S., Khoriba, G., and Ghoneim, A.S. (2021, January 27). OCFormer: A Transformer-Based Model For Arabic Handwritten Text Recognition. Proceedings of the International Mobile, Intelligent, and Ubiquitous Computing Conference (MIUCC), Cairo, Egypt.10.1109/MIUCC52538.2021.9447608 ↗
56. Momeni, S., and Babaali, B. (2022). Arabic Offline Handwritten Text Recognition with Transformers. Res. Sq.10.21203/rs.3.rs-2300065/v1 ↗
57. Generative Adversarial Network for an Improved Arabic Handwritten Characters Recognition
58. Generative adversarial network based adaptive data augmentation for handwritten Arabic text recognition10.7717/peerj-cs.861 ↗
59. Conditional Deep Convolutional Generative Adversarial Networks for Isolated Handwritten Arabic Character Generation10.1007/s13369-021-05796-0 ↗
60. Enhance to read better: A Multi-Task Adversarial Network for Handwritten Document Image Enhancement10.1016/j.patcog.2021.108370 ↗
61. Automatic Building of a Large Arabic Spelling Error Corpus10.1007/s42979-022-01499-x ↗
62. Deep Learning for Arabic Error Detection and Correction
63. Automatic Arabic Grammatical Error Correction based on Expectation-Maximization routing and target-bidirectional agreement10.1016/j.knosys.2022.108180 ↗
64. Abandah, G.A., Suyyagh, A., and Khedher, M.Z. (2021). Correcting Arabic Soft Spelling Mistakes using BiLSTM-based Machine Learning. arXiv.10.14569/IJACSA.2022.0130594 ↗
65. Synthetic data with neural machine translation for automatic correction in Arabic grammar10.1016/j.eij.2020.12.001 ↗
66. “Easy” meta-embedding for detecting and correcting semantic errors in Arabic documents10.1007/s11042-023-14553-4 ↗
67. Almajdoubah, A.N., Abandah, G.A., and Suvvagh, A.E. (2021, January 16–17). Investigating Recurrent Neural Networks for Diacritizing Arabic Text and Correcting Soft Spelling Mistakes. Proceedings of the IEEE Jordan International Joint Conference on Electrical Engineering and Information Technology (JEEIT), Dead Sea, Jordan.10.1109/JEEIT53412.2021.9634126 ↗
68. Irani, M., Elahimanesh, M.H., Ghafouri, A., and Bidgoli, B.M. (2022, January 28–29). A Supervised Deep Learning-based Approach for Bilingual Arabic and Persian Spell Correction. Proceedings of the 8th Iranian Conference on Signal Processing and Intelligent Systems (ICSPIS), Mazandaran, Iran.10.1109/ICSPIS56952.2022.10043875 ↗
69. Simple Extensible Deep Learning Model for Automatic Arabic Diacritization
70. Transfer learning and multi-phase training for accurate diacritization of Arabic poetry
71. Almanaseer, W., Alshraideh, M., and Alkadi, O. (2021). A Deep Belief Network Classification Approach for Automatic Diacritization of Arabic Text. Appl. Sci., 11.10.3390/app11115228 ↗
72. Self-information of radicals: A new clue for zero-shot Chinese character recognition10.1016/j.patcog.2023.109598 ↗
73. Puri, S. (2023). Advanced Network Technologies and Intelligent Computing, Springer.
74. Sharma, A., Katlaa, R., Kaur, G., and Jayagopi, D.B. (2023). Full-page handwriting recognition and automated essay scoring for in-the-wild essays. Multimed. Tools Appl.10.1007/s11042-023-14558-z ↗
75. Devanagari Text and Calligraphy Recognition Using ICF & ACF
76. Smys, S., Kamel, K.A., and Palanisamy, R. (2023). Inventive Computation and Information Technologies, Springer Nature.10.1007/978-981-19-7402-1 ↗
77. Sequence-aware multimodal page classification of Brazilian legal documents
78. Tanveer, M., Wang, Y., Amiri, A., and Zhang, H. (2023). DS-Fusion: Artistic Typography via Discriminated and Stylized Diffusion. arXiv.10.1109/ICCV51070.2023.00041 ↗
79. Fischer, N., Hartelt, A., and Puppe, F. (2023). Line-Level Layout Recognition of Historical Documents with Background Knowledge. Algorithms, 16.10.3390/a16030136 ↗
80. A survey on different feature extraction methods for writer identification and verification10.1504/IJAPR.2023.130511 ↗
81. Writer independent handwritten signature verification on multi-scripted signatures using hybrid CNN-BiLSTM: A novel approach10.1016/j.eswa.2022.119111 ↗
82. A Framework for Determining the Big Five Personality Traits Using Machine Learning Classification through Graphology
83. Chen, H.-J., Fathoni, H., Wang, Z.-Y., Lien, K.-Y., and Yang, C.-T. (2023). Smart Grid and Internet of Things, in Lecture Notes of the Institute for Computer Sciences, Social Informatics and Telecommunications Engineering, Springer Nature.