Real-time detection method and device for theft behaviors of climbing stairs and turning windows

Abstract

The invention discloses a real-time detection method and device for theft behaviors of climbing stairs and turning windows, and the method comprises the steps: constructing a YOLOv2-Res network whichis a network obtained by adding a residual module in the YOLOv2 network; sequentially inputting to-be-detected images in a to-be-detected image sequence into the trained YOLOv2-Res network, outputtinga plurality of candidate boxes and classification confidence coefficients corresponding to the candidate boxes, and obtaining the to-be-detected image sequence by framing the obtained monitoring video; screening the candidate boxes based on the classification confidence corresponding to the candidate boxes to obtain a target detection box; calculating a moving track of a target object in the to-be-detected image based on the target detection frame; matching the moving track of the target object with a preset stair climbing and window turning theft behavior, if the moving track conforms to thepreset stair climbing and window turning theft behavior, performing early warning, the stair climbing and window turning theft behavior can be detected in real time, and the situation that burglary occurs during stair climbing and window turning is prevented.

Description

technical field [0001] The present application relates to the technical field of computer vision, in particular to a method and device for real-time detection of theft by climbing stairs and turning over windows. Background technique [0002] At present, the behavior of burglary occurs frequently, and the common methods are climbing walls, turning windows or breaking doors, etc. Since it is difficult for people to find the behavior of climbing stairs and windows, after the perpetrator climbs stairs and windows to burglary, the victim can only After knowing that the property was stolen, in addition, when the police handling the case dealt with the burglary, they usually tracked down the suspect by checking the video surveillance, which made it difficult and inefficient to solve the case. Therefore, in order to severely crack down on the burglary behavior of climbing the building and turning the window, maintain social order and harmony, and save manpower and material resource...

AI Technical Summary

Problems solved by technology

[0002] At present, the behavior of burglary occurs frequently, and the common methods are climbing walls, turning windows or breaking doors, etc. Since it is difficult for people to find the behavior of climbing stairs and windows, after the perpetrator climbs stairs and windows

Images