Elder falling detection method based on deep learning, detection wristband and storage medium

Abstract

The invention provides an elder falling detection method based on deep learning, a detection wristband and a storage medium; body posture data and sign data are acquired by wearing the detection wristband, and the acquired data are calculated by using a constructed deep learning model. The method then includes correspondingly outputting a fall-down level according to the calculation result, outputting the fall-down level to the communication equipment, and giving an alarm. When the deep learning model is constructed, historical body posture data is learned through the neural network, conventional actions and habitual actions of the elderly under normal conditions can be effectively recognized, stress response when the elderly collides can be detected in time, and whether the elderly collides or not can be accurately judged by combining a set threshold value, missing report and wrong report are avoided. After the collision occurs, the invention includes evaluating the tumbling grade byusing the instantly detected sign data, and outputting an alarm. A rescuer is helped to know the body information of the old people in time and take correct rescue measures.

Description

technical field [0001] The invention relates to the field of smart wearables, in particular to a deep learning-based elderly fall detection method, a detection bracelet and a computer-readable storage medium. Background technique [0002] At present, my country's aging population is serious and has become an extremely serious social problem. If it is not dealt with correctly, it will seriously affect the development of my country's society, economy and other aspects. With the rapid development of social economy and the continuous improvement of people's living standards, the elderly have better pension conditions. At the same time, due to the advancement of modern medicine, the average life expectancy of the elderly has been greatly improved. Take Tianjin as an example: During the 10 years from 2007 to 2017, the population aging in Tianjin developed rapidly. The number of people aged 60 and above increased from 1.5629 million in 2007 to 2.4606 million in 2017. There are onl...

AI Technical Summary

Problems solved by technology

[0004] In the existing market, there are many devices worn by the elderly to detect falls of the elderly. Algorithms for calculating falls are installed in these devices, but the existing sensors for the existence of falls of the elderly are single, and the installed algorithms only pass The acceleration sensor detects the change of the old man's posture before and after the fall, and judges the fall. The algorithm is simple, and it is easy to miss or misjudgment. Moreover, there is no preliminary judgment on the physical condition of the old man after the fall. In the case of the consequences, it is inconvenient to rescue the elderly. Therefore, it is extremely necessary to have a detection method and detection equipment that can accurately judge the fall of the elderly and perform a preliminary evaluation of the fall level.

Images