Abstract
Arabic font recognition is the process of identifying the text font of the text images. Font recognition improves text recognition systems in terms of accuracy and time by selecting the appropriate font model for recognition. This paper presents an approach for recognizing Arabic text fonts on text images using Bag-of-Features technique. The proposed method is evaluated on two databases, APTI and KAFD. Comparable results are obtained. Using APTI database, a recognition rate of 99.9% is obtained while 96.46% and 98.71% recognition rates are obtained using 20 and 40 fonts of KAFD database, respectively.
| Original language | English |
|---|---|
| Title of host publication | IET Conference Proceedings |
| Publisher | Institution of Engineering and Technology |
| Pages | 628-632 |
| Number of pages | 5 |
| Volume | 2020 |
| Edition | 6 |
| ISBN (Electronic) | 9781839533303, 9781839534195, 9781839535062, 9781839535222, 9781839535239, 9781839535246, 9781839535406, 9781839535420, 9781839535635 |
| DOIs | |
| State | Published - 2020 |
Bibliographical note
Publisher Copyright:© 2020 The Institution of Engineering and Technology.
Keywords
- Arabic font recognition
- Bag-of-Features
- OCR
- Text recognition
ASJC Scopus subject areas
- General Engineering