Human falling detection method based on machine vision

Abstract

The invention provides a human falling detection method based on machine vision. Under the conditions that all detected objects are people in indoor environment and the activities of all people can be captured by a depth camera, the method comprises the following steps of: (1) acquiring an image by using the depth camera; (2) subtracting a background from the image which is acquired, and extracting a foreground from the image; (3) acquiring a foreground contour by using a binary image contour extraction algorithm; (4) performing convolution computation on the foreground contour and Gaussian functions in different scales, mapping a contour image to a curvature scale space (CSS) to form a CSS image, extracting peak points of the CSS image, and thus obtaining a video word bag based on curvature scale characteristics by the peak points in different scales; and (5) training an extreme learning machine classifier by using the video word bag which is obtained. By adoption of the method, under the condition of few training samples, an accurate falling detection result can be obtained within a short training time.

Description

technical field [0001] The invention relates to a method for detecting human falls by using machine vision, which belongs to the field of pattern recognition. Background technique [0002] The 21st century is called the "silver hair century", and the aging of the population is becoming a global development trend. In recent years, with the development of society and economy, the change of living style, the miniaturization of family structure and other factors, the empty nest rate of elderly families is constantly increasing. It is expected that by 2030, the proportion of empty-nest elderly families will reach 90%. At that time, the elderly families in my country will become seriously empty-nest. For the elderly, falls are a major health problem. A fall can cause serious injury and long-term immobility. About 50% of elderly people who require hospitalization are due to falls. Moreover, mortality from falls increases dramatically with age, with falls accounting for 57% and ...

AI Technical Summary

Problems solved by technology

"Monocular 3d head tracking to detect falls of elderly people" (based on monocular The 3D head tracking of the camera realizes the fall detection of the elderly) believes that the movement of the head has a great correlation with the fall of the human body, and proposes to use the 3D motion analysis algorithm of the head to detect the fall of the human body. This method requires detection, positioning and tracking head or face, time consuming

"Multiple object tracking for fall detection in real-time surveillance system" (used in real-time monitoring system for human Multiple Object Tracking for Fall Detection) points out that monocular camera solutions are limited by occlusions when tracking multiple fall events

"Human body posture classification by a neural fuzzy network and home" published in "IEEE Transactions on Systems, Man and Cybernetics: Part A" (System, Man and Cybernetics IEEE Journal Section A) at 37 (6): 984-994 in 2007 "care system application" (Human posture classification based on fuzzy neural network and its application in home care) uses fuzzy neural network to detect whether a fall has occurred, which requires a large number of training samples and a long training time

Due to the characteristics of classifier training, the detected object needs to be consistent with the training object, so when the size and posture of the person are seriously inconsistent with the training sample, normal classification and detection cannot be performed.

Images