Journal of American Science, 2012;8(3) http://www.americanscience.org http://www.americanscience.org editor@americanscience.org 133 Real-time Crowd Monitoring using Infrared Thermal Video Sequences Adnan Ghazi Abuarafah 1 ; Mohamed Osama Khozium 2 and Essam AbdRabou 2 1 Vice Dean for Academic Affairs, Faculty of Computer of and IS Makkah, Umm Al- Qura University. Makkah, SA 2 Center of Research Excellence in Hajj and Omrah (HajjCore) Makkah, Umm Al- Qura University, Makkah, SA agabuarafah@uqu.edu.sa ,osama@khozium.com Abstract:Monitoring people in a crowded environment is a critical task in civilian surveillance. Most vision-based counting techniques depend on detecting individuals in order to count their number. Counting becomes inefficient when it is required in real-time and when the crowd is dense. This paper proposes a novel technique for monitoring and estimating the density of crowd in real-time using infrared thermal video sequences. The research targets monitoring the crowd in Muslims’ pilgrimage event (Hajj) while almost 3.0 million Muslims gather in Makkah to perform Hajj. During different Hajj phases the movement of the gathered Muslims is required at the same time from a place to another. Thus monitoring their crowd in real-time is crucial in order to take immediate decisions to prevent crowd disasters. A state of the art thermal camera has been acquired for the surveillance process. In addition, special software modules have been developed to analyze the captured thermographic video sequences in real-time. The results show high accuracy of the estimation of the crowd density in real-time. [Adnan Ghazi Abuarafah;Mohamed Osama Khoziumand EssamAbdRabouReal-time Crowd Monitoring using Infrared Thermal Video Sequences]Journal of American Science 2012; 8(3): 133-140].(ISSN: 1545-1003). http://www.americanscience.org . Keywords: Hajj, Islamic informatics, crowd density estimation, crowd monitoring, computer image understanding, Thermography, computer vision. 1. Introduction Around the 9 th day of Hajj month, more than 3.0 million Muslims gather in Makkah to perform Hajj. This number will continue to increase in the next few years to reach almost 3.75 million Muslims by the year 2019[1]. Moving this giant number of people with uncontrolled manner resulted in many accidents in the past twenty years [2]. For example, in 2004, 251 died in a stampede at Mena Valley in Makkah[2]. Avoiding critical crowd densities and triggering rapid group movement in a relatively short period of time have become critical tasks[3] required to avoid future accidents and keeping the sanctity of emotions at its best. So providing real-time information about the density and behavior of the crowd in a certain place or route is the aim of this research. This paper proposes a computer vision methodology for estimating of crowd density. Crowd density is defined as the number of people in the audience relative to the facility’s capacity[3]. The target is to integrate the proposed technique with an intelligent decision support system for crowd management during Hajj rituals. The proposed technique uses a far infrared camera (sometimes also called forward looking infrared or FLIR camera) for monitoring and estimating the density of crowd in real- time. Although far infrared cameras were originally developed for military use, decreasing their prices make them now available for civilian use. There are some applications of it in industry especially in quality control. However, its technology is not yet widely available, and many researchers lack the experience in acquiring far infrared (thermographic) images[4]. The proposed technique analyzes video sequences captured by a FLIR camera and calculates the occupied area of land in the captured scenes. Then, the crowd density of that area can be calculated. The analyzing algorithm works in real-time by continuously analyzing the camera output. A graph can be plotted to show the crowd density of a certain area against time. From this graph, the behavior of the crowd whether it is accelerating or decelerating can be determined. The results of the proposed technique are validated in two experiments which show high accuracy of providing crowd density and behavior estimation in real-time. This paper is arranged in six sections. Section one is this introduction, section two discusses infrared Thermography from a theoretical point of view for Thermography. Section 3 lists some related work. Section four describes the methodology. Section five discusses the experimental work. Finally, section 6 concludes the work and discusses the future of the research. Infrared Thermography Thermographic images represent the electromagnetic radiation of an object in the far infrared range, which is 6 − 15µm. The principle of Thermography is based on the physical phenomenon that any body of a temperature above absolute zero (-273.15 °C) emits electromagnetic radiation. The electromagnetic spectrum is the range frequencies of electromagnetic radiation and is divided, lowest