Abnormal behavior detection method and system based on human skeleton, and medium

Abstract

The invention provides an abnormal behavior detection method and system based on a human skeleton, and a medium, and the method comprises the steps: S1, building a training video sample database, andbuilding an original video database with a preset frame length according to a public violent behavior database; s2, according to the obtained original video database, extracting human body skeletons in all video samples in the original database by utilizing an open-source attitude estimation algorithm OpenPose to obtain a skeleton video database; s3, deframing all the obtained video samples in theskeleton video database into a skeleton picture stream by using an FFmpeg tool to obtain a skeleton picture stream; S4, sequentially inputting the obtained skeleton picture streams into a 3D-DenseNetnetwork model for training, so as to obtain a trained 3D-DenseNet network model. The human skeleton and the 3D-DenseNet are used for abnormal behavior detection for the first time, the detection accuracy is high, the robustness is high, and the detection result can be output and updated in real time.

Description

technical field [0001] The present invention relates to the technical field of image detection, in particular, to a method, system and medium for abnormal behavior detection based on a human skeleton. In particular, it relates to an intelligent abnormal behavior detection method based on human skeleton and 3D-DenseNet network model. Background technique [0002] In recent years, with the large-scale deployment and use of surveillance systems, a large amount of surveillance video data has been generated. The role of the monitoring system is to detect target behavior and abnormal behavior, and it becomes increasingly difficult to screen and detect videos only manually. Therefore, relying on the method of computer vision to analyze the surveillance video, so as to automatically detect a series of violent behaviors that endanger public safety such as fights and fights that appear in the surveillance video, is of great significance for maintaining social stability and realizing ...

AI Technical Summary

Problems solved by technology

This method proposes to use the human skeleton for abnormal behavior detection, but the method of calculating the human skeleton uses a top-down "frame difference method"

In the crowded state of many people, the time overhead is large and the skeleton of each person cannot be well distinguished, and the robustness is poor

Images