Method and device for detecting tumbling

Abstract

The invention provides a method and device for detecting tumbling. The method provided by the invention comprises the steps of: firstly, collecting behavioral data of a user; secondly, identifying behavior of the user according to the behavioral data of the user; thirdly, segmenting out behavioral switching data from the collected data according to a behavioral identification result, and warping the behavioral switching data into equilong characteristic vectors; and fourthly, performing tumbling detection according to the warped characteristic vectors. According to the method and device provided by the invention, tumbling detection is performed on the basis of behavioral switching, a great amount of normal behavioral data can be filtered, the complexity of a feature space is reduced, the separating capacity of models is enhanced, and the model detection rate is improved.

Description

technical field [0001] The invention relates to the fields of pervasive computing and health monitoring, in particular to a fall detection method and device. Background technique [0002] According to statistics from the World Organization for Disease Control and Prevention, about one-third of the home-based elderly over 65 years old fall more than once a year, and nearly 10% of one-time falls will cause serious injury or disease. Falls have become one of the important factors that endanger the elderly and other special populations. Timely fall detection and rescue can win valuable time for treatment and rescue, and play a very important role in improving the independent living ability of users, ensuring their health and improving the level of medical monitoring. At present, my country has entered an aging society and is showing a trend of accelerated development. The elderly living alone and "empty-nest families" are gradually increasing. There is an urgent need to develop...

AI Technical Summary

Problems solved by technology

Due to the high similarity between the instantaneous process of individual normal behaviors (such as running, going down stairs, etc.) Balance between rates: The threshold method needs to artificially set the threshold size, and the discrimination boundary defined by the model method is also affected by related parameters, it is difficult to find the optimal discrimination boundary between normal behavior and abnormal behavior (falling)

Therefore, although there are many methods for fall detection, none of the existing methods can meet the requirements of high detection rate and low false alarm rate for fall detection at the same time.

However, fall detection is a cost-sensitive problem. On the one hand, the consequences of missed detection of falls are much more serious than those of normal behaviors, requiring the model to have a high detection rate; on the other hand, frequent false alarms will cause The disgust of users reduces their trust in the detection device, which is not conducive to the actual application and promotion of the device

Images