Development of an Unobtrusive Sleep Monitoring System using a Depth Sensor

Abstract

The elderly, who are isolated and neglected, can experience to an increased likelihood of age-related sleep disorders. Early detection, proper diagnosis, and treatments for sleep disorders can be achieved by identifying sleep behaviours through long-term monitoring. Although many studies developed sleep monitoring systems by using non-invasive measures, research is still limited to detect sleep behaviours using an infrared 3D depth camera. The present study has two main goals: a) identifying design characteristics on the existing sleep monitoring systems based on the extensive literature review; b) developing a non-invasive sleep monitoring system using a 3D depth camera. For the literature review, several journal papers have been reviewed to identify the design features, classification accuracy, and embedded algorithms of the existing sleep monitoring systems. For the system development, a sleep posture monitoring system was developed, using a Microsoft Kinect II depth sensor. For the validation of the developed system, the differences of sleep behaviours between the two participants were investigated. The developed system used an unobtrusive measure and it would be useful for long-term monitoring on sleep behaviours for the senior population.