Raspberry Pi elder people tumbling detection system based on deep learning

Abstract

The invention discloses a Raspberry Pi elder people tumbling detection system based on deep learning. Each collector comprises two parallel three-axis acceleration sensors, and the collectors are placed on a waistband, pockets of trousers and pockets of a coat of a person needing collection respectively. When changes of acceleration and angular velocity exceed a certain range, alarm information issent out; redundant information is contained in the data, the data is reduced from 150 dimensions to 100 dimensions, then the data of behavior states is trained through a DBN deep neural network, anda network model generated through training is used for detecting the real-time behavior states. The system has the advantages that the tumbling detection principle is based on the DBN deep neural network, and false tumbling can be effectively prevented, for example, when equipment freely falls on the ground, mistaken judgment of false tumbling is avoided. The cost of Raspberry Pie hardware equipment is low, and popularization can be better achieved.

Description

technical field [0001] The invention belongs to the technical field of electronic devices, and relates to a raspberry pie elderly fall detection system based on deep learning. Background technique [0002] The growing population of the elderly has become the focus of attention. Due to the inconvenient physical activities of the elderly, falls have become the fourth cause of casualties in my country, and accidental falls are the main health threat to people over 65 years old. Relevant scholars in my country have conducted research on the fall problem of the elderly, but the products produced by the research institute are mainly crutches and walkers. Although these products can reduce the probability of the elderly falling, they cannot effectively rescue the elderly when they fall in the first place. Therefore, this study is based on the combination of traditional products and the application of the Internet of Things era, and explores its results on the elderly who will not...

AI Technical Summary

Problems solved by technology

At present, although some fall detection based on MEMS technology in China can better realize fall detection, most of them have a large amount of calculation, complex design, and high price, so it is difficult to be widely promoted.

Images