Fall-down and Smile Recognition for Elderly Adults Using Kinect
碩士 === 國立高雄應用科技大學 === 電機工程系博碩士班 === 104 === This study proposes a system which can help elderly adults contact their family immediately when they fall. We proposed installing the device in home areas regularly occupied by elderly adults and identifying the type of their fall by detecting their skele...
| Main Authors: | , |
|---|---|
| Other Authors: | |
| Format: | Others |
| Language: | zh-TW |
| Published: |
2016
|
| Online Access: | http://ndltd.ncl.edu.tw/handle/dthwaa |
| Summary: | 碩士 === 國立高雄應用科技大學 === 電機工程系博碩士班 === 104 === This study proposes a system which can help elderly adults contact their family immediately when they fall. We proposed installing the device in home areas regularly occupied by elderly adults and identifying the type of their fall by detecting their skeleton posture. When falls are detected, text messages were immediately sent to the mobile phones of family members to notify them of the emergency and to show them live images via a webpage. In addition, a smile detection system is developed to recognize the smile intensity from detecting the image of mouth corners. By adequately arranging the RGB weightings for the image gray-scale of mouth region in a face, mouth-corner is identified effectively. The width and height of the mouth-region image is analyzed to identify the mouth shape and the smile intensity. Four smile scales are proposed in this study. This enabled the device to learn about their everyday emotions and can serve as a reference to family members or community care centers when caring for elderly adults.
|
|---|