Crowd Behaviour Analysis based Automated Surveillance System

Abstract

The country is experiencing turbulence in the global environment hence stabilizing the nation is a challenge as it seeks to address other issues such as security in Pakistan. To address this problem, we have embarked on developing a solution that aims at enabling the user to quickly detect acts of violence among crowds before they degenerate into full-blown altercations. In our proposed solutions, the framework is highly integrated by merging cultural understanding and the experience of the user in a digital platform that tackles the limitations of the current solutions. This system is intended to assist the identified authorities, as well as protect the public through the effective use of live CCTV feeds, thoughtful analysis of recorded footage, and intelligent security database management. In doing so, our hope is to not only raise security standards but enhance safety for everyone within their respective facilities. The foundation of our approach consists of two primary components: Convolutional Neural Networks (CNN) and Long Short-Term Memory (LSTM) elements. This high-end technology exhibits high effectiveness in identifying abusive behavior much higher than existing systems with an overall efficiency of 86%. Through our use of technology, our aim is not only to provide a secure environment for the society but also to help everyone to have a better living standard. We believe that our approach represents a significant step forward in achieving this goal.