| Summary: | 碩士 === 長庚大學 === 資訊工程學系 === 98 === Taiwan’s elderly population is being increased year by year. Today the fall accidents have become one of important causes of death forelderly. More and more elder communities have installed high-securityvideo surveillance system to insure the safety for elders. Thesesurveillance systems, however, almost need employ guards or care-staffs to watch the monitors artificially. Some high-and systems apply the imageprocessing techniques to automatically identify if there are exceptional events appeared in the monitors. Such systems, however, will have performance bottlenecks on image processing. In addition, the convention video surveillance systems lack better emergence alarming mechanism in case an elder falling incident occurs.
For this reason, we will adopt a hierarchical framework to implement a falling detection system with distributed image processing and data flow control. Through distributed image processing, it will effectively reduce the burden on Server. Also, the Server will perform data flow control forthe data stream collected by all cameras based on Round-Robin principles. The most important is, through the combination of mobile applicationsscreening on highrisk factors of the diseases, and personnel positioningmechanism , the system will support more complete services for elder safety.
With the development of this system, we make the fall-accidentdetection from occurring to alarming in a very short delay time . And wewill take into account physiological conditions for seperate elderly to give more accurate , personal services. Our goal is to hope the elderly can do activities more confidently and safely in the community.
|
|---|
