Video abnormal event detection method and chip

Abstract

The invention relates to a video abnormal event detection method and a chip. The method comprises the following steps: acquiring a to-be-detected video; using the to-be-detected video to obtain at least two to-be-detected images arranged according to a time sequence; determining the spatial similarity between the adjacent to-be-detected images to obtain the spatial change feature of the to-be-detected video, wherein the spatial change feature comprises a change rule of spatial similarity according to a time sequence; determining the time sequence similarity between the adjacent to-be-detected images, and obtaining the time sequence change feature of the to-be-detected video, wherein the time sequence change feature comprises a change rule of the time sequence similarity according to a time sequence; fusing the spatial change feature and the time sequence change feature to obtain a fused change feature; utilizing a pre-trained detection model to determine whether the fusion change feature is abnormal or not; and if so, determining that an abnormal event exists in the to-be-detected video. According to the scheme, the video abnormal event identification efficiency can be improved.

Description

technical field [0001] The invention relates to the technical field of image processing, in particular to a video abnormal event method and chip. Background technique [0002] With the continuous development of Internet of Things technology, monitoring equipment has been widely deployed in major public areas, providing people with implicit security guarantees. However, traditional video surveillance mainly relies on human monitoring of abnormal events in the scene, which not only requires extremely high labor costs, but is also susceptible to subjective factors, and may even lead to some abnormal events not being discovered in time, making the abnormalities in the existing surveillance video Event recognition is less efficient. [0003] In view of this, in view of the above deficiencies, it is necessary to provide a video abnormal event detection method and chip to solve the above deficiencies. Contents of the invention [0004] The technical problem to be solved by the ...

AI Technical Summary

Problems solved by technology

However, traditional video surveillance mainly relies on human monitoring of abnormal events in the scene, which not only requires extremely high labor costs, but is also susceptible to subjective factors, and may even lead to some abnormal events not being discovered in time, making the abnormalities in the existing surveillance video Event recognition is less efficient

Images