Video abnormity detection method based on graph structure under multi-scale transformation

Abstract

The invention discloses a video abnormity detection method based on a graph structure under multi-scale transformation. First of all, video abnormity detection is carried out by use of an optical flow feature, and spatial correlation scales of video abnormity association optical flow features in different scenes are different. Under the condition that the spatial correlation structure of the video abnormity association optical flow features is maintained, since the calculation efficiency of video abnormity detection can be effectively improved by reducing the quantity of the optical flow features, it is brought forward that a network graph structure of the optical flow features is constructed based on spatial correlation, by use of multi-scale transformation of the graph structure under correlation constraints, the quantity of the optical flow features in the video abnormity detection is effectively reduced, under the condition that the detection precision is slightly reduced, the calculation efficiency of a video abnormity detection algorithm can be substantially improved, and detection of event abnormities is more rapid and reasonable.

Description

technical field [0001] The invention belongs to the technical field of video anomaly detection, and in particular relates to a video anomaly detection method based on multi-scale transformation of the lower image structure. Background technique [0002] In recent years, the problem of public safety has become increasingly prominent. The abnormal events of crowds in public places are detected in time and relevant departments are notified for response and rescue, so as to reduce the personal casualties and property losses of the masses. Therefore, video-based crowd anomaly detection is particularly important. In traditional video anomaly detection techniques, the usually simplified social dynamic model uses optical flow features combined with latent Dirichlet distribution to complete anomaly detection, but this method fails to fully express the motion information. For this reason, some people have proposed that based on the defects of the large amount of optical flow calculat...

AI Technical Summary

Problems solved by technology

Someone proposed a method of mixing dynamic textures for detection, but the implementation of mixed dynamic texture features is relatively cumbersome. When modeling with multiple texture models, it is necessary to use the contrast information between the front and back frames of the texture, and the complexity of this method and high time-consuming affect the fast detection characteristics of anomaly detection

Although this detection method has a certain effect in the anomaly detection of video moving objects, the inability to realize anomaly detection in real time also leads to the failure of effective promotion of this method.

Images