Al-Masjid An-Nabawi Crowd Adviser Crowd Level Estimation Using Head Detection
Raghad Jaza Alamri, Maha Suliman Alkhuriji, Malak Saed Alshamani, Omniyyah Yahya Ibrahim, Fazilah Haron · 2018
Crowd disasters are common in places of human gathering. Thousands of people gather at AI-Masjid An-Nabawi to perform their daily prayers. The mosque is packed with people and finding the right entrance is always an issue. The purpose of this study is to identify the most recent method in computer vision that could be used to estimate the level of crowdedness at these entrances based on the existing CCTV videos. The method will then be incorporated in a mobile application, which acts as a “crowd adviser” to direct the visitors of AI-Masjid An-Nabawi to less crowded doors (if any) using indoor built map. The system detects the heads in the crowd using Aggregate Channel Features (or Viola-Jones depending on the performance), count the detected objects to estimate the crowd level and present it to the the user on a mobile application.