A close-loop verification approach for pedestrian stability based on machine vision

In public places, it is significant to analyze the stability of crowd which can support management and control, protect evacuees safely effectively. The numerical analysis method system based on Lyapunov theory suffers problems that difficult avoid random errors in initialization pedestrian density velocity, as well cumulative due time increasing, limiting its application. This study adopts a complementary model theoretical machine vision with parallel convolutional neural network (CNN) model. It proposes an approach closed-loop verification for merging systems. Thereby, this research provides methodological planning functional layout flow crowd-gathering places control measures stable flow.