Method for carrying out turbulent detection through video monitoring

Abstract

The invention provides a method for carrying out turbulent detection through video monitoring. The method comprises the following steps of: firstly, determining a velocity field of crowd movement by calculating optical flow between two adjacent frames of images, and carrying out smooth filtering on the velocity field so as to provide reliable microscopic movement information for mesoscopic small group clustering; furthermore, density clustering is carried out by adopting a DBSCAN algorithm, crowd behaviors are divided into different behavior types, motion modes with adjacent positions and small speed difference are combined into a small group type, and a microscopic motion field is positioned to a mesoscopic layer to be analyzed; and finally, dividing the velocity field characteristic space into a normal region and an abnormal region according to the velocity, regarding the category of which the clustering center falls in the abnormal region and the retention time is relatively long asan abnormal behavior, and mapping velocity field coordinates to an image coordinate system to determine an abnormal region.

Description

technical field [0001] The invention belongs to the technical field of public security video monitoring, and relates to a method for finding out whether there is a riot phenomenon in a crowd through monitoring video. Background technique [0002] Usually, crowd riots will cause bad social impact, seriously endanger personal safety and social order, and may even threaten social stability and national security. Crowd riots generally refer to chaotic scenes that cannot be controlled and stopped in a short period of time due to the overcrowding of crowds gathered somewhere during a certain event or activity. In order to avoid potential riots, measures such as public security control and safety inspection are used. However, traditional manual supervision often has problems such as regulatory loopholes and poor timeliness. Sometimes because of crowds and large numbers of people, it will cause secondary damage. cause serious consequences. Contents of the invention [0003] The ...

AI Technical Summary

Problems solved by technology

In order to avoid potential riots, measures such as public security control and safety inspection are used. However, traditional manual supervision often has problems such as regulatory loopholes and poor timeliness. Sometimes because of crowds and large numbers of people, it will cause secondary damage. cause serious consequences

Images