Falling abnormal behavior identification method based on law enforcement and case handling area scene

Abstract

The invention belongs to the technical field of abnormal behavior analysis, and particularly relates to a falling abnormal behavior identification method based on a law enforcement and case handling area scene, and the method can be divided into four parts of two-dimensional image data acquisition, human body detection based on deep learning, human body key point positioning and falling behavior identification. The method comprises the following steps: (1) acquiring video image data in a law enforcement and case handling scene through a camera; (2) obtaining a coordinate position of a human body detection frame through a target detection network; (3) obtaining the position of each key point of the human body by using a human body key point detection network for each human body frame, and (4) inputting each human body detection frame into a classification network for dichotomy to judge whether a person falls to the ground, thereby improving identification effect and identification precision.

Description

technical field [0001] The invention belongs to the technical field of abnormal behavior analysis, and in particular relates to a method for identifying abnormal behavior of falling to the ground based on the scene of a law enforcement case handling area. Background technique [0002] Existing technology: Aiming at the recognition method of falling behavior in the scene of law enforcement and case handling area, a high recognition rate is required to prevent the life-threatening situation of people falling to the ground in the activity area. However, the existing judgment process of falling behavior recognition is generally: (1) acquisition of image data; (2) image data processing to obtain human body image or its features; (3) feature classification to identify falling behavior. [0003] For different steps, there are different methods for identifying down behavior. In the first stage, the two-dimensional or three-dimensional image data of the human body can be obtained th...

AI Technical Summary

Problems solved by technology

And for the scene in the law enforcement case handling area, when the person is at the edge of the image or bends over or sits on the ground and only part of the human body is detected, the existing technology is prone to false detection and cannot accurately identify the shortcomings of the falling behavior

Images