IAES International Journal of Artificial Intelligence (IJ-AI) Vol., No., 2012, pp. xx~xx ISSN: 2252-8938  1 Journal homepage: http://iaesjournal.com/online/index.php/IJAI Counting of People in the Extremely Dense Crowd using Genetic Algorithm and Blobs Counting Muhammad Arif [1,2] , Sultan Daud [1,2] , Saleh Basalamah [1,2] [1] Center of Research Excellence in Hajj and Omrah (HajjCoRE), Umm Al-Qura University, Makkah, Saudi Arabia [2] College of Computer and Information Systems, Umm Al-Qura University, Makkah, Saudi Arabia Article Info ABSTRACT (10 PT) Article history: Received Nov 13 th , 2012 Revised Accepted In this paper, we have proposed a framework to count the moving person in the video automatically in a very dense crowd situation. Median filter is used to segment the foreground from the background and blob analysis is done to count the people in the current frame. Optimization of different parameters is done by using genetic algorithm. This framework is used to count the people in the video recorded in the mattaf area where different crowd densities can be observed. An overall people counting accuracy of more than 96% is obtained. Keyword: People Counting Image Processing Blob Analysis Median filtering Genetic algorithm Copyright © 201x Institute of Advanced Engineering and Science. All rights reserved. Corresponding Author: Muhammad Arif, College of Computer and Information Systems, Umm Al-Qura University, Makkah, Saudi Arabia. Email: mahamid@uqu.edu.sa 1. INTRODUCTION (10 PT) Crowd monitoring is very important in many aspects especially in the areas of Airports, railway stations, sports, and rallies. In Saudi Arabia, Hajj and Ramadan are the mega events in Makkah Mukarama where lot of people get together to do umra and hajj. Crowd management during Ramadan and Hajj at the Holy Mosque in Makkah is a daunting task. Tremendous effort from the security staff is required to manage the huge crowd peacefully and smoothly. In the last decade, due to low cost of cameras, lots of cameras were used for the surveillance of public places. Lot of cameras are installed in the Grand Mosque that help the security staff to take appropriate crowd management decisions. Normally uneven crowding occurs in the Mosque due to tendency of people entering in the nearest possible entry points when they arrive in the Mosque. This situation creates suffocation in these areas and becomes prone to certain mishap or loss of precious lives. Looking at these installed cameras and estimating the crowd density and directing the people to empty places requires huge amount of manpower. Manual analysis of high quantity of visual data is not practical and an automatic decision support system is required that can guide the security staff and public to minimize the crowding of people in certain places and optimize the usage of the grand mosque. Lot of research is being done in automating the process of estimation and management of crowd using visual cameras, thermal imaging or other sensors placed at the entry points. In the last decade, due to low cost of cameras, lots of cameras were used for the surveillance of public places. Manual monitoring of crowd is done by putting many surveillance cameras and some observers monitor the crowd density and their movement. This scenario is very costly as lot of manpower is required who can watch the monitors continuously for many hours. Hence, alertness of the observers is an important factor in good surveillance. As the working hours increases as the case of Masjid-e-Haram, fatigue and stress of the observer increases degrading their performance. The importance and demand for automated tools to manage and analyze crowd behavior and dynamics grows day by day as the population increases. People counting in