Arabic Handwritten Text Recognition Systems and Challenges and Opportunities

Document Type : Original Article

Authors

1 Computer Science, Faculty of Computer and Artificial Intelligence, Damietta University, Damietta, Egypt

2 information system department, faculty of computer and artificial intelligence, Damietta University

3 Department 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.

Keywords

The Egyptian Journal of Language Engineering
Volume 10, Issue 2
October 2023
Pages 84-103

Files

Share

How to cite

Statistics