Counting People in a Crowd Using Viola-Jones Algorithm
International journal of computing, communication and instrumentation engineering · 2017
Counting the number of people in crowded areas such as the Masa'a of Al-Masjid Al-Haram has become a necessity.This is due to the increasing number of people performing Hajj or Umrah.In this paper, we propose a crowd counting system for the Masa'a using image processing techniques.The system detects the number of faces using the Viola-Jones detector then count the number of detected objects.The results show that the accuracy of the algorithm depends on the crowd density level, which are 77%, 82% and 79%, for low, medium and high density, respectively.The relatively low accuracy was due to the women wearing niqab, the face cover (usually black) since it is difficult to distinguish between a covered face from the back of the head.The system keeps the count results (i.e. total people heading to Safa or Marwah) in a database and presents the stored information in a histogram so that the authority is able to study the pattern of crowd flow and in case of overflow the system shows an alert window to warn the authorities of this dangerous situation in order to effectively manage the crowd at the Masa'a.