Text Line Segmentation of Al-Quran Pages Using Binary Representation

One of the early tasks in a handwriting recognition system is the segmentation of a handwritten document that is imaged into text lines, which is the process of defining the region of every text line on a document’s image. In this paper, the focus is in the text-line segmentation of mushaf holy Quran. Thus, Segmentation of the mushaf holy Quran is based on the processes of segmenting the Arabic/Jawi handwritten texts. Any incorrect segmentation will affect the holiness of the Quran. In this study, a novel algorithm was proposed for detecting the text line on the Quran pages. The research aim is to segment the Quran pages to text line images without any changes on the content. Firstly, pre-processing that was applied includes binarization. Then, the Quran pages were segmented to the text line. The experimental results on several of the Quran pages from different Al-Quran styles demonstrate the effectiveness of the proposed technique.