Sleep monitoring system based on body posture movement using Microsoft Kinect sensor

Sleep activity is one of important factor to determine quality of human life and closely related with sleep quality, which is influenced by several factors such as daily activities, stress, and fatigue. Current sleep monitoring device is called polysomnography, which is commonly used on sleep monito...

Ausführliche Beschreibung

Gespeichert in:
Bibliographische Detailangaben
Hauptverfasser: Febriana, Nityacas, Rizal, Achmad, Susanto, Erwin
Format: Tagungsbericht
Sprache:eng
Schlagworte:
Online-Zugang:Volltext
Tags: Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
Beschreibung
Zusammenfassung:Sleep activity is one of important factor to determine quality of human life and closely related with sleep quality, which is influenced by several factors such as daily activities, stress, and fatigue. Current sleep monitoring device is called polysomnography, which is commonly used on sleep monitoring systems in hospital by attaching several electrodes to the subject’s head. However, physical contacts between device and subject may leads to disruption of sleep activity because subject feels uncomfortable. Goal of this study is to design a sleep monitoring system using Microsoft Kinect v.2. This device is able to track movements made by the subject. The subject’s sleep quality is determined by the number of movements during bedtime. Joints displacement in time unit is calculated using the Euclidean distance method to calculate activities during sleep. The sleep monitoring system is designed to recognize subject’s dominant sleep posture by using the boundary location method to divide captured body into three parts that separated with three baseline algorithms. Analysis results of sleep monitoring system output are classification of subject’s sleep quality and the dominant sleep posture. Designed system is tested for 105 minutes and subject’s posture changes per minute are called Minutely Posture Movement (MPM). Sleep quality is classified into 3 categories, namely “Good”, “Normal”, and “Bad”. The classification constants of “Good”, “Normal”, and “Bad” are obtained from Q1 and Q3 of 10 subjects MPM. Value Q1 is 0.08 and value Q3 is 0.15. Subject’s sleep quality is categorized as follows: “Good” on MPM
ISSN:0094-243X
1551-7616
DOI:10.1063/1.5096680