Behavior detection method and device, electronic equipment and computer readable storage medium

Abstract

The invention provides a behavior detection method and device, electronic equipment and a computer readable storage medium. The method comprises the following steps: extracting a plurality of image frames from a monitoring video according to a time sequence; inputting the plurality of image frames into a pedestrian detection model, detecting whether a pedestrian exists in each image frame in the plurality of image frames by the pedestrian detection model, and generating a pedestrian frame in the image frame in which the pedestrian exists; when the pedestrians exist in at least two image frames in the plurality of image frames, determining displacement pixel distances of the pedestrians in adjacent image frames through a preset target tracking algorithm; when the displacement pixel distance is larger than or equal to a preset threshold value corresponding to the size of the pedestrian frame, the image areas, with the pedestrian frame, in the multiple image frames are input into a posture detection model, a detection result which is output by the posture detection model and represents whether the pedestrian runs or not is obtained, and therefore the accuracy and reliability of behavior detection can be improved.

Description

technical field [0001] The present application relates to the technical field of computer vision, and in particular, to a behavior detection method, device, electronic equipment, and computer-readable storage medium. Background technique [0002] In some specific scenarios, there is a need to detect the running behavior of people. For example, if there are people running in some parks, it often indicates that there may be an emergency. Timely detection of abnormal running behavior of personnel can help supervisors quickly locate abnormal event status, troubleshoot on-site risk factors, and ensure the safety of other people's lives and property. At present, manual inspection or traditional machine learning methods are usually used to detect people's running behavior. Park video surveillance systems often monitor tens or hundreds of channels of video at the same time. Due to the limitations of human vision, manual inspection methods often miss situations, and this method has...

AI Technical Summary

Problems solved by technology

[0003] The purpose of this application is to provide a behavior detection method, device, electronic equipment and computer-readable storage medium, which can improve the problems of missed detection and false positives in the detection of running behavior in complex monitoring scenarios

Images