Signal Processing for Non-intrusive Sleep Monitoring
| Project/Area Number |
15K12072
|
|---|
| Research Category |
Grant-in-Aid for Challenging Exploratory Research
|
| Allocation Type | Multi-year Fund |
|---|
| Research Field |
Perceptual information processing
|
|---|
| Research Institution | National Institute of Informatics |
|---|
Principal Investigator |
CHEUNG GENE 国立情報学研究所, コンテンツ科学研究系, 准教授 (40577467)
|
|---|
| Co-Investigator(Kenkyū-buntansha) |
小野 順貴 首都大学東京, システムデザイン研究科, 教授 (80334259)
|
|---|
| Project Period (FY) |
2015-04-01 – 2018-03-31
|
| Project Status |
Completed (Fiscal Year 2017)
|
| Budget Amount *help |
¥3,640,000 (Direct Cost: ¥2,800,000、Indirect Cost: ¥840,000)
Fiscal Year 2016: ¥1,820,000 (Direct Cost: ¥1,400,000、Indirect Cost: ¥420,000)
Fiscal Year 2015: ¥1,820,000 (Direct Cost: ¥1,400,000、Indirect Cost: ¥420,000)
|
|---|
| Keywords | 画像処理 / グラフ信号処理 / sleep monitoring / graph signal processing / 奥行き動画処理 / 画像復元 / 機械学習 |
|---|
| Outline of Final Research Achievements |
The goal of this research is to monitor the sleep quality of a patient non-intrusively via video and audio recording and analysis. In particular, we focus on apnea detection, a common and serious sleep condition that affects a large percentage of the Japanese older population. The recorded video is a sequence of depth images captured by a Microsoft Kinect camera, which utilizes active sensing technologies, so that captured images are not affected by ambient lighting conditions. The captured video is noise-corrupted and of low bit-depth, and requires denoising and bit-depth enhancement, performed using graph-signal restoration techniques. Features from video and audio are then extracted for supervised learning to construct a classifier. The designed classifier can then detect different apnea types with high accuracy, and is robust to the patient's sleep pose. The prototype has been deployed in an Australian sleep clinic and has demonstrated its effectiveness.
|
Report
(4 results)
Research Products
(9 results)
