El-Awadly, E., Ebada, A., Al-Zoghby, A. (2023). Arabic Handwritten Text Recognition Systems and Challenges and Opportunities. The Egyptian Journal of Language Engineering, 10(2), 84-103. doi: 10.21608/ejle.2023.193993.1043
Esraa Mohamed K. El-Awadly; Ahmed Ismael Ebada; Aya Mohammed Al-Zoghby. "Arabic Handwritten Text Recognition Systems and Challenges and Opportunities". The Egyptian Journal of Language Engineering, 10, 2, 2023, 84-103. doi: 10.21608/ejle.2023.193993.1043
El-Awadly, E., Ebada, A., Al-Zoghby, A. (2023). 'Arabic Handwritten Text Recognition Systems and Challenges and Opportunities', The Egyptian Journal of Language Engineering, 10(2), pp. 84-103. doi: 10.21608/ejle.2023.193993.1043
El-Awadly, E., Ebada, A., Al-Zoghby, A. Arabic Handwritten Text Recognition Systems and Challenges and Opportunities. The Egyptian Journal of Language Engineering, 2023; 10(2): 84-103. doi: 10.21608/ejle.2023.193993.1043
Arabic Handwritten Text Recognition Systems and Challenges and Opportunities
1Computer Science, Faculty of Computer and Artificial Intelligence, Damietta University, Damietta, Egypt
2information system department, faculty of computer and artificial intelligence, Damietta University
3Department of Computer Science, Faculty of Computers and Information Science Damietta University Damietta, Egypt
Abstract
Arabic handwritten text recognition faces significant challenges despite the large number of Arabic speakers. A critical review paper has analyzed previous research in this field, identifying problem areas and challenges faced by researchers. The paper focuses on trends in offline handwriting recognition systems and the unique characteristics of the Arabic language that pose technical challenges. The analysis involved comparing and contrasting previous research methods and performances to summarize critical problems and enumerate issues that must be addressed. The paper highlights several Arabic datasets that can be utilized as benchmarks for training, testing, and comparisons. These datasets are essential for evaluating the performance of Arabic handwriting recognition systems. Additionally, the paper concludes with a fundamental comparison and discussion of remaining open problems and trends in the field. It identifies several unresolved technical issues, such as the need for improved feature extraction and modeling techniques, as well as the need for large-scale, diverse datasets to facilitate better training and testing of Arabic handwriting recognition systems. Overall, the paper provides a comprehensive overview of the challenges and issues facing Arabic handwriting recognition and highlights areas where further research is needed.
View on SCiNiTO
Top