Capsule Embedded ResNet for Arabic Handwriting Character Recognition
DOI:
https://doi.org/10.22399/ijcesen.3626Keywords:
Handwriting , Arabic characters recognition, Characters recognition, Residual Network, Capsule NetworksAbstract
This paper presents novel hybrid architectures for Arabic handwritten character recognition, integrating capsule networks with residual neural networks (ResNets) across various embedding strategies. The proposed Custom Caps-ResNet models explore low-, mid-, high-, and multilevel capsule embeddings to synergize hierarchical feature learning with spatial relationship preservation. Evaluated on four benchmark datasets, the models achieve competitive accuracy—99.64% on OIHACDB-28 and 94.14% on Dhad—while consistently reducing test loss by up to 80% on Dhad and 66% on HMBD_V1 compared to baselines. These reductions in loss indicate enhanced prediction certainty and improved feature representation. Multilevel and mid-level embeddings perform robustly across diverse script complexities, whereas high-level embeddings excel in semantic abstraction. The variation in dataset performance reveals how capsule networks mitigate challenges in cursive connections, overlaps, and positional character forms. Overall, the integration of capsule embeddings into ResNet hierarchies leads to not only strong accuracy but also significantly more confident predictions—advancing Arabic handwriting recognition toward reliable real-world deployment.
References
[1] Impedovo, S., Wang, P. S. P., & Bunke, H. (1997). Automatic bankcheck processing (Vol. 28). World Scientific.
[2] Grabowski, H. (2023). Intelligent character recognition of handwritten forms with deep neural networks. In International TRIZ Future Conference (pp. 3–15). Springer Nature Switzerland. https://doi.org/10.1007/978-3-031-44768-61
[3] Louloudis, G., Gatos, B., Pratikakis, I., & Halatsis, C. (2008). Text line detection in handwritten documents. Pattern Recognition, 41(12), 3758–3772. https://doi.org/10.1016/j.patcog.2008.05.011
[4] Doermann, D. (1998). The indexing and retrieval of document images: A survey. Computer Vision and Image Understanding, 70(3), 287–298. https://doi.org/10.1006/cviu.1998.0695
[5] Kaur, H., & Kumar, M. (2023). Signature identification and verification techniques: State-of-the-art work. Journal of Ambient Intelligence and Humanized Computing, 14(2), 1027–1045. https://doi.org/10.1007/s12652-021-03503-7
[6] Filatov, A., Gitis, A., Kil, I., & Nikandrov, A. (1998). The address script recognition system for handwritten envelopes. In International Workshop on Document Analysis Systems (pp. 767–774). Springer Berlin Heidelberg. https://doi.org/10.1007/3-540-69709-8_77
[7] Eberhard, D. M., Simons, G. F., & Fennig, C. D. (Eds.). (2024). Ethnologue: Languages of the world (27th ed.). SIL International. https://www.ethnologue.com/
[8] Lorigo, L. M., & Govindaraju, V. (2006). Offline Arabic handwriting recognition: A survey. IEEE Transactions on Pattern Analysis and Machine Intelligence, 28(5), 712–724. https://doi.org/10.1109/TPAMI.2006.102
[9] Miloud, K., Benkhedda, Y., & Barkat, A. (2024). Restoration of ancient Arabic manuscripts: A deep learning approach. Studies in Engineering and Exact Sciences, 5(2), e7722. https://doi.org/10.54021/seesv5n2-033
[10] Alothman, A., Al-Turaiki, I., AlGhamdi, A. D., AlKhulaiwi, D., AlHassan, R., & AlOmran, H. (2025). A survey on the online Arabic handwriting recognition: Challenges, datasets, and future directions. In Seventeenth International Conference on Machine Vision (ICMV 2024) (Vol. 13517, pp. 1–12). SPIE. https://doi.org/10.1117/12.3057143
[11] Alzubaidi, L., Zhang, J., Humaidi, A. J., Al-Dujaili, A., Duan, Y., Al-Shamma, O., Santamaría, J., Fadhel, M. A., Al-Amidie, M., & Farhan, L. (2023). A survey on deep learning tools dealing with data scarcity: Definitions, challenges, solutions, tips, and applications. Journal of Big Data, 10(1), 46. https://doi.org/10.1186/s40537-023-00727-2
[12] Altwaijry, N., & Al-Turaiki, I. (2021). Arabic handwriting recognition system using convolutional neural network. Neural Computing and Applications, 33(7), 2249–2261. https://doi.org/10.1007/s00521-020-05070-8
[13] Alwagdani, M. S., & Jaha, E. S. (2023). Deep learning-based child handwritten Arabic character recognition and handwriting discrimination. Sensors, 23(15), 6774. https://doi.org/10.3390/s23156774
[14] Kamal, M., Shaiara, F., Abdullah, C. M., Ahmed, S., Ahmed, T., & Kabir, M. H. (2022). Huruf: An application for Arabic handwritten character recognition using deep learning. In 2022 25th International Conference on Computer and Information Technology (ICCIT) (pp. 1131–1136). IEEE. https://doi.org/10.1109/ICCIT57492.2022.10055758
[15] Balaha, H. M., Ali, H. A., Saraya, M., & Badawy, M. (2021). A new Arabic handwritten character recognition deep learning system (AHCR-DLS). Neural Computing and Applications, 33(11), 6325–6367. https://doi.org/10.1007/s00521-020-05397-2
[16] Ullah, Z., & Jamjoom, M. (2022). An intelligent approach for Arabic handwritten letter recognition using convolutional neural network. PeerJ Computer Science, 8, e995. https://doi.org/10.7717/peerj-cs.995
[17] Alghyaline, S. (2024). Optimised CNN architectures for handwritten Arabic character recognition. Computers, Materials & Continua, 79(3), 4907–4924. https://doi.org/10.32604/cmc.2024.048734
[18] Najam, R., & Faizullah, S. (2023). Analysis of recent deep learning techniques for Arabic handwritten-text OCR and post-OCR correction. Applied Sciences, 13(13), 7568. https://doi.org/10.3390/app13137568
[19] AlShehri, H. (2024). DeepAHR: A deep neural network approach for recognizing Arabic handwritten recognition. Neural Computing and Applications, 36(21), 12103–12115. https://doi.org/10.1007/s00521-024-09674-2
[20] AlMuhaideb, S., Altwaijry, N., AlGhamdy, A. D., AlKhulaiwi, D., AlHassan, R., AlOmran, H., & AlSalem, A. M. (2024). Dhad—A children’s handwritten Arabic characters dataset for automated recognition. Applied Sciences, 14(6), 2332. https://doi.org/10.3390/app14062332
[21] Torki, M., Hussein, M. E., Elsallamy, A., Fayyaz, M., & Yaser, S. (2014). Window-based descriptors for Arabic handwritten alphabet recognition: A comparative study on a novel dataset. arXiv preprint arXiv:1411.3519. https://arxiv.org/abs/1411.3519
[22] Loey, M., El-Sawy, A., & El-Bakry, H. (2017). Deep learning autoencoder approach for handwritten Arabic digits recognition. arXiv preprint arXiv:1706.06720. https://arxiv.org/abs/1706.06720
[23] Ghofrani, A., & Toroghi, R. M. (2019). Capsule-based Persian/Arabic robust handwritten digit recognition using EM routing. In 2019 4th International Conference on Pattern Recognition and Image Analysis (IPRIA) (pp. 168–172). IEEE. https://doi.org/10.1109/PRIA.2019.8785981
[24] Al-Taani, A. T., & Ahmad, S. (2021). Recognition of Arabic handwritten characters using residual neural networks. Jordanian Journal of Computers and Information Technology, 7(2), 192–205. https://doi.org/10.5455/jjcit.71-1615204606
[25] Lutf, M., You, X., Cheung, Y. M., & Chen, C. P. (2014). Arabic font recognition based on diacritics features. Pattern Recognition, 47(2), 672–684. https://doi.org/10.1016/j.patcog.2013.09.029
[26] Kada, B., Mohammed, A., & Abdelmajid, B. (2025). An optimized approach for handwritten Arabic character recognition based on the SVM classifier. Engineering, Technology & Applied Science Research, 15(2), 22232–22238. https://doi.org/10.48084/etasr.9292
[27] Cheriet, M., Al-Badr, B., & Suen, C. Y. (2012). A robust word spotting system for historical Arabic manuscripts. In Guide to OCR for Arabic scripts (pp. 453–484). Springer London. https://doi.org/10.1007/978-1-4471-4072-6_18
[28] El-Sawy, A., Loey, M., & El-Bakry, H. (2017). Arabic handwritten characters recognition using convolutional neural network. WSEAS Transactions on Computer Research, 5, 11–19.
[29] Yousfi, S., Berrani, S. A., & Garcia, C. (2015). ALIF: A dataset for Arabic embedded text recognition in TV broadcast. In 2015 13th International Conference on Document Analysis and Recognition (ICDAR) (pp. 1221–1225). IEEE. https://doi.org/10.1109/ICDAR.2015.7333958
[30] Al-Ohali, Y., Cheriet, M., & Suen, C. Y. (2009). Arabic handwriting recognition using baseline-dependent features and hidden Markov modeling. Pattern Recognition, 42(10), 2854–2869. https://doi.org/10.1016/j.patcog.2009.04.003
[31] Abed Hussein, B., & Malik, L. (2020). An overview on Arabic handwriting recognition and database. International Journal of Electrical & Computer Engineering, 10(1), 105–114. https://doi.org/10.11591/ijece.v10i1.pp105-114
[32] Mesleh, A. M., & Khorsheed, M. S. (2022). A deep learning approach for Arabic handwriting recognition: A survey. IEEE Access, 10, 53215–53227. https://doi.org/10.1109/ACCESS.2022.3177818
[33] Liu, W., Chen, W., Wang, C., Mao, Q., & Dai, X. (2021). Capsule embedded ResNet for image classification. In Proceedings of the 2021 5th International Conference on Computer Science and Artificial Intelligence (pp. 143–149). ACM. https://doi.org/10.1145/3507548.3507571
[34] Kamal, A. (2024). Enhancing lung cancer detection through a novel CapsuleNet-ResNet fusion model: A comparative study on accuracy and robustness. International Journal of Intelligent Systems and Applications in Engineering, 12(3), 4357–4365.
[35] Allenki, J., & Soni, H. K. (2024). Enhancing liver tumour segmentation in CT images using dilated residual capsule networks. Journal Européen des Systèmes Automatisés, 57(6), 1775–1783. https://doi.org/10.18280/jesa.570625
[36] Sabour, S., Frosst, N., & Hinton, G. E. (2017). Dynamic routing between capsules. arXiv preprint arXiv:1710.09829. https://arxiv.org/abs/1710.09829
[37] Boufenar, C., Kerboua, A., & Batouche, M. (2018). Investigation on deep learning for off-line handwritten Arabic character recognition. Cognitive Systems Research, 50, 180–195. https://doi.org/10.1016/j.cogsys.2017.11.002
[38] Boufenar, C., & Batouche, M. (2017). Investigation on deep learning for off-line handwritten Arabic character recognition using Theano research platform. In 2017 Intelligent Systems and Computer Vision (ISCV) (pp. 1–6). IEEE. https://doi.org/10.1109/ISACV.2017.8054968
[39] Balaha, H. M., Ali, H. A., Youssef, E. K., Elsayed, A. E., Samak, R. A., Abdelhaleem, M. S., ... & Mohammed, M. M. (2021). Recognizing Arabic handwritten characters using deep learning and genetic algorithms. Multimedia Tools and Applications, 80(21), 32473–32509. https://doi.org/10.1007/s11042-021-11103-2
[40] Boufenar, C., & Batouche, M. (2016). OIHACDB: A new database for offline isolated handwritten Arabic character recognition. In Actes des Posters du Treizième Colloque sur l’Optimisation et les Systèmes d’Information-COSI’2016 (p. 20). Université Ferhat Abbas Sétif.
Downloads
Published
How to Cite
Issue
Section
License
Copyright (c) 2025 International Journal of Computational and Experimental Science and Engineering
This work is licensed under a Creative Commons Attribution 4.0 International License.
