Framework of Page Segmentation for Mushaf Al-Quran Based on Multiphase Level Segmentation

Abstract

This paper presents the framework of page segmentation for Mushaf Al-Quran based on Multiphase Level Segmentation (MLS). This study focuses to (a) extract multiform frame shape by using a novel technique Neighbouring Pixel Behaviors (NPB) and (b) segment text line by using a novel technique which is Hybrid Projection Based Neighbouring Properties (HPBNP). Since Mushaf Al-Quran pages are decorated with a different type of pattern and design of a decorative frame. Thus, the decoration frame must be properly to extract out from a page of Mushaf Al-Quran first before properly get only the text of Mushaf Al-Quran regardless of its decoration heterogeneity. Therefore, NPB technique was proposed to remove multiform frame shape from the page of Mushaf Al-Quran. While the text of Mushaf Al-Quran has a several of diacritical marks, hence it will block the process of segmenting text line. Therefore, HPBNP technique was proposed for segment overlapping text line that interfered by diacritical marks or the stroke of the Arabic word. Experimental results of the proposed technique is shown in this paper.