Many-people abnormal behavior identification method based on human body shape characteristic

Abstract

The invention discloses a many-people abnormal behavior identification method based on a human body shape characteristic. The method comprises the following steps of rapidly carrying out target human body detection, using a lot of pedestrian sample gradient information to make a series of target human body detection filters so as to acquire a pedestrian detector of a group of filters, calculating a characteristic channel of an image to be detected, using the detector to calculate a characteristic response on each channel, and finally using an Adaboost classifier to acquire a final target human body position; using the human body shape characteristic and prior information of the acquired human body position, constructing an initial human body appearance model, and adopting an appearance transmission mechanism; using a Hoff shape characteristic extraction algorithm to extract a characteristic of each part of a human body from the appearance model and matching with an abnormal behavior sample database so as to acquire a final identification result. By using the method provided in the invention, under the condition that a background environment and a camera position are not certain, position calibration of many people can be realized and final abnormal behavior identification are realized too.

Description

technical field [0001] The invention belongs to the technical field of image processing, and relates to a method for identifying abnormal behaviors of multiple human bodies. Specifically, it is a system for identifying abnormal behaviors of multiple people based on the morphological features of human bodies by extracting the morphological features of each human body in a video image. Background technique [0002] The Intelligent Visual Internet of Things (IVIOT) is an important part of the new generation of information technology and an upgraded version of the Internet of Things. The intelligent visual Internet of Things is to perceive people, vehicles, and objects through visual sensors, information transmission, and intelligent visual analysis. According to the agreed agreement, any object is connected to the Internet for information exchange and communication, so as to realize the intelligent recognition of objects. , location tracking and real-time monitoring of an intel...

AI Technical Summary

Problems solved by technology

[0006] The technical problem to be solved by the present invention is: Aiming at the difficulty in identifying abnormal human behaviors and the low accuracy of simultaneous identification by multiple people, a method for identifying abnormal human behaviors is proposed to realize high-precision abnormal behavior identification

Images