Deep Learning Based Suspicious Activity Detection in Mass Gathering using HAAR cascade Algorithm

  • Unique Paper ID: 151031
  • Volume: 7
  • Issue: 11
  • PageNo: 399-405
  • Abstract:
  • Human activity recognition from real-time video is an efficient area of research for image processing and computer vision. With visual employment, human activities can be viewed in sensitive and public areas such as bus stations, railway stations, airports, banks, shopping malls, schools and colleges, car parks, roads, etc. Preventing terrorism, theft, accidents and illegal parking, vandalism, fighting, chain exploitation, crime and other suspicious activities were done. It is very difficult to view public places on an ongoing basis, so there is a need for intelligent video surveillance that can monitor people's activities in real time and distinguish them from normal and unusual activities; and may issue a warning. This Project provides precautionary measures based on the use of suspicious activity that can be used on surveillance cameras. Real-time cameras are becoming increasingly popular and are widely distributed in homes, offices, and public areas. As the number of camera views increases, one camera operator is unable to control the entire monitoring process and is limited by human resources. Also, it is not possible for humans to ensure continuous monitoring at all times. The project proposes an Intelligent Surveillance System based on in-depth study that can alert a person's operator or relevant authority to take appropriate action by posting using the IMAP protocol (Internet Message Access Protocol) where suspicious activity, disrupt public order and unusual human behavior such as administration is detected other weapons such as a gun, knife or glass bottles, which can be very dangerous. In this project, in order to allow for real use of the system, we focus on reducing the number of false alarms. Our system therefore ensures public safety where many people gather to avoid unnecessary consequences.

Cite This Article

  • ISSN: 2349-6002
  • Volume: 7
  • Issue: 11
  • PageNo: 399-405

Deep Learning Based Suspicious Activity Detection in Mass Gathering using HAAR cascade Algorithm

Related Articles