Many methods have been proposed for Arabic font recognition, but none of them has considered the specialty of the Arabic writing system. Most of these methods are either general pattern recognition approaches or application of other methods which have been developed for languages other than Arabic. Therefore, this paper is the first attempt to present an alternative method for Arabic font recognition based on diacritics. It presents the diacritics as the thumb of Arabic fonts which can be used individually to identify and recognize the font type. Diacritics are the marks and strokes which have been added to the original Arabic alphabet. Though they are the smallest regions in the Arabic script, with today technology it is very easy to get a high resolution image with a very low cost. In this kind of images, the diacritics can reveal very useful information about the font type. In this study, two algorithms for diacritics segmentation have been developed, namely flood-fill based and clustering based algorithm. The experiments conducted proved that our approach can achieve an average recognition rate of 98.73% on a typical database that contains 10 of the most popular Arabic fonts. Compared with existing methods, our approach has the minimum computation cost and it can be integrated with OCR systems very easily. Moreover, it could recognize the font type regardless of the amount of the input data since five diacritics, which in most cases can be found in only one word, are enough for font recognition.
Scopus Subject Areas
- Signal Processing
- Computer Vision and Pattern Recognition
- Artificial Intelligence
- Arabic diacritics
- Composite of central and ring projection
- Font recognition