Fall detection method and system for housebound old people based on multi-feature fusion

Abstract

The invention discloses a fall detection method and system for housebound old people based on multi-feature fusion and belongs to the technical field of computer vision. The method and system can effectively monitor the potential tumble risk of the housebound old people and improve the pre-judgment accuracy of the potential tumble risk of the housebound old people. The method comprises the following steps of: carrying out real-time video acquisition on a given monitoring object, and obtaining voice signals and video signals; extracting acoustic features of the voice signals; acquiring current posture features and face features of the monitored object, acquiring a current heart rate value of the monitored object based on the face features, and acquiring a posture response peak value of the monitored object based on the current posture features; obtaining expression features of the monitored object; evaluating the feature confidence of the current posture features, the expression features and the current heart rate value of the monitored object to determine the fusion weight of each feature, and obtaining the fusion fall confidence of the monitored object; and judging the falling condition of the monitored object based on the fusion falling confidence and the posture response peak value of the monitored object.

Description

technical field [0001] The invention belongs to the technical field of computer vision, and in particular relates to a fall detection method and system for the elderly at home based on multi-feature fusion. Background technique [0002] With the acceleration of the informatization process, the smart home system has also ushered in new changes, and it has also had a relatively positive impact on video surveillance in the home. In the process of safety monitoring and analysis of the elderly at home, a very important task is to analyze the potential fall risk of the elderly. As one of the behaviors that can have a direct impact on the human body, falling will not only have a direct negative impact on human health, but also bring potential health risks. Falls are the number one cause of injury for people aged 65 and over, the data show. About 9,500 older adults die from falls each year, and the average person aged 65 to 69 suffers a hip fracture from every 200 falls. More ser...

AI Technical Summary

Problems solved by technology

This method takes a long time, and it is difficult to balance speed and accuracy

Although the above methods can achieve a certain recognition effect in a certain scene, it is difficult for a single feature to fully represent the rich information of the action of falling, and the recognition rate in complex scenes needs to be improved, and considering the flexibility of recognition Therefore, it is necessary to provide a fall behavior recognition method based on multi-feature fusion, so as to achieve better results in home security.

Images