A Sleeping Quality Analysis System Based onThermal Images

• Main Authors: Chi-You Ou, 歐季祐
• Other Authors: Shaou-Gang Miaou
• Format: Others
• Language: zh-TW
• Published: 2010
• Online Access: http://ndltd.ncl.edu.tw/handle/19909239731178332281

碩士 === 中原大學 === 電子工程研究所 === 98 === Sleeping is the most fundamental activity of human life, and approximately one third of our life is spent on it. When we do not have enough sleep, it will indirectly affect our daily activities. People in Taiwan are gradually paying attention to the problem of sleeping quality in recent years. Thus a number of sleep monitoring centers are established, where they provide sleep monitoring and analysis services to help people solve this problem. However, their service requires sufficient manpower and sophisticated, expensive equipments to maintain certain levels of service quality. This thesis proposes an easy operating sleeping quality analysis system based on thermal images. The proposed system can be applied to both home and health-care institutions. Using thermal images has the advantage of non-contact. Moreover, it has lower privacy concern and can obtain some important physiological information (such as body surface temperature) that cannot be obtained by using traditional images. The proposed system adopts Motion History Imaging (MHI) to record motion energy and track breathing behavior of human. By breathing through the nostrils to produce temperature variation, we obtain corresponding thermal images. The thermal images are then converted into a one-dimensional and near periodic breathing signal for the calculation of its frequency. Finally, a pair of thresholds is selected for the thresholding of motion energy and the standard deviation of breathing frequency in order to obtain the information on sleeping status. In addition, the proposed system is compared to a breathing and motion detection device carried at wrist. Experimental results show that the proposed system can track the human breathing behavior in real time, and resolve the issue where a subject without any movement should be determined as sleeping or awake. Finally, the result of sleeping quality analysis is close to the actual value, showing that the proposed system is effective.