Automatic Extraction of Tents during Hajj from Airborne Images
Dr. Abdulatif Alharthy
College of Engineering and Islamic Architecture, Umm Al-Qura University
Makkah, Saudi Arabia
Email: alharthy@uqu.edu.sa
Dr. Nabeel Koshak
Design and Planning Support Systesms Unit,
Hajj Research Institute, Umm Al-Qura University, Makkah, Saudi Arabia
Email: n@cad-gis.com
Abstract With the huge number of pilgrims performing Hajj these years, urban space utilization is a major concern for engineers, and urban designers/planners. Pilgrims must stay in the Holy areas (Arafat and Mina) for few days as part of Hajj rituals. Given these criteria tents as light weight temporary structures are used to accommodate pilgrims.
The extensive demand for rapid, automatic, and high quality algorithms for feature extractions has been the subject for many researches recently. In this paper, we present an approach to detect and extract tent using airborne images. The approach basically is used to calculate the areas covered by tents. It utilizes the intensity in digital images in two stages. First, it classifies tents from other features. Second, it calculates the number of tents based on image matching subroutines.
Using this automatic approach, number of pilgrims can be estimated of a tested area given the average capacity of one meter square covered by tents. Based on such results urban space utilization can be testified. Moreover, services, utilities, and transportation needs can be determined more precisely. An actual sample area in Arafat during the Hajj season is used to testify the approach developed through this research.