Fall Detection Using a Single Video Camera
1 online resource (61 pages) : PDF
University of North Carolina at Charlotte
Falls among the elderly are concerning, especially if they go unnoticed for an extended period of time. Due to fear of falling, many people end up in assisted living facilities or nursing homes which are costly and the lack of independence greatly affects quality of life. Finding ways to allow the elderly to safely stay in their own homes longer not only reduces cost of living and increases well-being, but also alleviates the families’ burden of worrying about their loved ones. Due to their low cost, cameras are a convenient option for monitoring. This study made use of this and developed an algorithm to be used with a single video camera in order to determine when someone has fallen or is about to fall, so an alarm could be set off to alert a caretaker for help. The developed algorithm was tested for a variety of scenarios including falling down stairs, partially occluding the subject, low light levels, and placing the camera at a variety of angles. The acceleration values obtained were also compared with an Inertial Measurement Unit. The algorithm successfully tracked the person and detected falls in each scenario. The comparison with the IMU sensor points toward further potential for clinical applications in fall risk assessment.
Tabarraei, AlirezaSmelser, Ronald
Thesis (M.S.)--University of North Carolina at Charlotte, 2015.
This Item is protected by copyright and/or related rights. You are free to use this Item in any way that is permitted by the copyright and related rights legislation that applies to your use. For other uses you need to obtain permission from the rights-holder(s). For additional information, see http://rightsstatements.org/page/InC/1.0/.
Copyright is held by the author unless otherwise indicated.