Disturbance Propagation Model of Turn-Back Behavior in a Crowd Flow Network and Elimination Mechanism Approach

The ability to recognize unexpected turn-back behavior in a crowd flow network is a significant development in the rapidly growing area of pedestrian behavior detection in public security. In this paper, we proposed a disturbance propagation model (DPM) using a convolutional neural network (CNN) based on visual geometry group (VGG) architecture to detect pedestrian turn-back behavior in a crowd flow network. The proposed strategy looks for turn-back movements, a key sign of addressing crowd disturbances to prevent crowd disasters. An approach to eliminate mechanisms is proposed, which involves identifying non-intrusive alerts such as light or silent sound. This study highlights how technology can improve public safety in densely populated areas and offers considerable insights into active crowd management strategies.