Fall detection method based on key points

Abstract

The invention discloses a fall detection method based on key points, a fall detection system based on key points comprises an image acquisition module, an image recognition module, an action detectionmodule and a display module which are connected in sequence, the image acquisition module comprises a camera, and the camera acquires images in an area frame by frame; the image recognition module detects a human body in the continuous frame images and recognizes skeleton key points of the detected human body; the action detection module detects a linear distance formed by a wrist key point and an ankle key point in the human skeleton key points in the continuous frame images and a linear distance formed by an eye key point and the ankle key point based on an aspect ratio that the target detection frame conforms to a set threshold value; judges whether a human body in the video is in a falling state or not according to the change condition of the included angle between the shoulder key points and the knee key points; and the display module is used for displaying whether a person is in a falling state in the video image area through a display and giving a corresponding warning prompt.

Description

technical field [0001] The invention belongs to the technical field of motion detection, and relates to a fall detection method based on key points. Background technique [0002] Accidental falls and injuries often occur, especially in nursing homes where there are many elderly people, or construction sites that require high-intensity physical labor. The harm caused by accidental falls is often very serious, and it is possible to lack timely detection and treatment life-threatening. At present, the research objects of researchers on fall detection technology are often the elderly, but the fall detection methods and applications of workers in construction sites are still a blank field. The working environment of construction workers is harsh, and the work intensity is high. During the work process, they often fall or even faint. If they cannot be treated in time, the consequences are often very serious. [0003] According to the different sensing methods, the existing fall ...

AI Technical Summary

Problems solved by technology

[0019] 1) The existing detection methods are aimed at single person detection, and cannot realize simultaneous multi-person fall detection;

[0020] 2) Each fall judgment is based on key point judgment, lacking timing control;

Images