A sleep posture and sleep behavior testing and recognition method without disturbing sleep

Abstract

A non-sleep-interference sleeping posture and sleeping behavior testing identification method includes the steps of extraction of human-bed interface compression shapes, classification and identification of sleeping postures, classification and identification of sleeping motion behaviors and analysis of sleeping behavior modes. A support vector machine is used for classification and identification of the sleeping postures, and the four sleeping postures of supination, left lateral decubitus, right lateral decubitus and prostration are identified on the basis of finite sample supervised training identification through classification of human-bed interface compression quantity matrixes; identification of the sleeping motion behaviors is achieved in the mode that logarithmic potential sensors synchronizing with human-bed interface compression quantity changes are used for outputting variation of voltage signals and obtaining signals generated when sleeper motion behavior accidents occur. According to the method, a technical support is provided for monitoring and evaluation of intellectualized design of a bed system or a mattress and healthy sleeping.

Description

technical field [0001] The invention relates to an automatic recognition technology of sleeping posture and sleeping behavior. Background technique [0002] The bedding system or mattress is often considered to be an important external factor affecting sleep quality. During sleep, the bedding system or mattress keeps in contact with the human body, supports the human body, relaxes the muscles of the human body and restores the intervertebral disc. The ideal bedding system or mattress support conditions should make the user's prone spine in a naturally curved state. It is based on this that many advanced mattresses are currently designed as "zone-adjustable", that is, through the adjustment of the hardness characteristics of the mattress's shoulders, waist, buttocks, etc., it responds to the different spinal support needs of people of different sizes. But in fact, a large number of experimental studies have shown that this kind of mattress does not achieve the purpose of sig...

AI Technical Summary

Problems solved by technology

Although the polysomnography recording method can obtain detailed information of sleep structure, this method can only be completed in the laboratory, and polyconductive electrodes need to be placed on the human body, so it has a certain degree of influence on normal sleep

Actigraphy analysis method is a better method for testing specific human body movements, but in order to obtain comprehensive sleep action behaviors, it is necessary to wear actigraphy devices on multiple parts of the human body, so it will also affect sleep. a certain degree of influence

Although the camera method can test sleeping posture and sleep behavior without disturbing sleep, this method does not respect the privacy of the subjects enough, and at the same time, it cannot accurately test subtle movement behaviors and movement behaviors covered by bedding

In addition, since these test methods cannot link the sleeping position and sleeping behavior with the mechanical properties of the bed or mattress, it is also impossible to intelligently change the mechanical properties of the bed or mattress according to the requirements of sleeping position and sleeping behavior

Images