Text Line Segmentation for Mushaf Al-Quran Using Hybrid Projection Based Neighbouring Properties
Amirul Ramzani Radzid, Mohd Sanusi Azmi, Intan Ermahani A. Jalil, Nur Atikah Arbain, Anas Tahir · Journal of Telecommunication Electronic and Computer Engineering (JTEC) · 2018
Text line segmentation is an important step in document image processing. Its part of the pre-processing stage to prepared the images before throughout either feature extraction or classification images. In this paper, we present a method of line segmentation for Mushaf Al-Quran text using a hybrid projection based neighbouring properties. This is based on the pixel, object and histogram properties. This method will identify overlaps between neighbouring text lines and segment each line with precision. Overlap caused by interfering with diacritical marks or stroke of the Arabic word must be properly segmented without change the original meaning of the text. Experimental results show the validity of our method.