Farhad Tehranipour - Department of Electrical and Electronic Engineering University of Melbourne, Melbourne, Australia
Rosita Shishegar - Department of Electrical and Electronic Engineering University of Melbourne, Melbourne, Australia
Soheil Tehranipour - Department of Electrical and Computer Engineering Qazvin Branch, Azad University, Qazvin, Iran
Kamaledin Setarehdan - School of Electrical & Computer Engineering University of Tehran, Tehran, Iran

one important issue in machine vision is usingautomatic attention control methods for monitoring CCTVcameras, in order to enhance the security of people in public.Result of automatic methods such as crowd density estimationcan alert the operator in the case of risk probability increasing.In addition to overall crowd density, other parameters such asregional crowd density and the temporal and spatial criteria ofeach frame of video should be considered to control theoperator’s attention correctly. For this purpose, according to thegradual change of crowd density and risk probability in dailyhours and uncertainty in our knowledge in evaluation of crowdedplaces, we designed a fuzzy decision making system to makedecisions about risk probability. The design of this system isbased on the fact that the human visual system tends to directattention to events that happen with low probability. Theefficiency of this system is tested on real data and results arepresented to demonstrate the practical applications of this systemto aid the human operator.

Crowd density estimation, fuzzy decision-making Attention control, CCTV cameras,wavelet transform,support vector machine