Algorithms to Detect In-Home Falls of Elderly Using Structured Light Sensing

点击翻译按钮获取中文摘要

Abstract Each year more than one out of four people 65 and older falls. Of those that fall one out of five has a serious injury such as broken bones or a head injury. Over 800,000 patients a year are hospitalized because of a fall injury, most often because of a head injury or hip fracture. Studies have found an increased risk of physical and physiological complications associated with prolonged periods of lying on the floor following a fall, due to an inability to get up. Older adults living alone are at great risk of delayed assistance following a fall. A low- cost, unobtrusive system capable of automatically detecting falls in the homes of older adults could help significantly reduce the incidence of delayed assistance after a fall. This project will develop an innovative new in-home fall monitoring system that solves many practical problems with other systems. The technical approach uses structured light. Structured light sensing creates 3D point clouds of a scene to allow detection of motion sequences which will allow for the automatic detection of a person’s fall. There are many benefits of this approach; the elderly person is not required to carry an electronic device or take action after the fall, video images that would create a privacy concern for the elderly person are not used, the system can work in darkness unlike camera based approaches, and the system is less susceptible to false alarms than vibration or acoustic fall monitoring that infer a fall. The structured light sensing fall detection product is intended to be an optional accessory to in-home alert systems in use today. It will work with multiple vendors of in-home alert systems. It will operate in parallel with wearable buttons to signal an alert. Once the structured light sensing fall detection identifies a fall it will activate the alert system the same way as a wearable button press. The same protocol would be followed. In these systems an operator would first try to talk to the person with the speaker phone of the vendors’ in-home alert system. If they cannot communicate with the person, they start working through a call list of local people to check on the home. The structured light sensing fall detection system would be used if caregivers decide for various reasons that a simple wearable button is not an adequate solution for the person being monitored.