Crowd aggregation feature analysis method and system for intelligent security

Abstract

The invention discloses a crowd aggregation feature analysis method for intelligent security, and the method comprises the following steps: S1, extracting a moving individual target from a crowd videoimage of security monitoring, and analyzing the movement behavior features of the moving individual target; S2, carrying out architecture analysis on the crowd according to the motion behavior characteristics of the motion individual target. Based on the method, a corresponding system is designed, the motion behavior characteristics of the motion individual target are combined with the architecture of the crowd; through utilizing crowd structuring degrees, the state of the crowd is judged, and the complex and tedious steps of the crowd overall architecture analysis in the prior art are avoided, so the accuracy of an analysis result is improved, and the efficiency of crowd architecture analysis is also improved; different levels of index monitoring can be adopted for different types of crowds according to the crowd structuralization degree, so the index monitoring has pertinence and accuracy.

Description

technical field [0001] The invention relates to the technical field of video security monitoring, and more specifically relates to a crowd aggregation feature analysis method and system for smart security. Background technique [0002] For public spaces such as stations, squares, and building corridors where crowds gather, the focus of video security monitoring should not be limited to specific individuals, such as people with abnormal behaviors such as retrograde, stagnation, and falling to the ground. Attention should also be paid to the crowd formed by the aggregation of a certain number of specific individuals; these specific individuals may be co-travelers or strangers, but they gather together due to the proximity of space during the movement process to form a crowd; In public spaces, the phenomenon of crowd aggregation is very common. Therefore, analyzing the characteristics of crowd aggregation is very important to maintain the safety and smooth traffic of public spa...

AI Technical Summary

Problems solved by technology

[0003] However, analyzing the aggregation characteristics of the crowd is much more complicated than analyzing the behavior characteristics of specific individuals. First, the behavior of individuals in the crowd is both affected by the crowd and reacts on the crowd. Serious anomalies such as serious abnormalities are often caused by the behavior of one or a few individuals first, but the relationship between individuals and the crowd and the impact on the crowd are very diverse. Under what circumstances will the behavior of one or a few individuals cause the entire It is difficult to judge the abnormality of the crowd and under what circumstances will not obviously affect the crowd; secondly, there are obvious differences in the appearance of the normal behavior and abnormal behavior of the individual, and the external appearance of the crowd in various states is relatively different when they appear. Large anomalies are very similar, and it is difficult to carry out effective prediction and classification

Images