【摘要】 随着老龄化进程的加剧，基于智能健康设计的居住环境有望解决社会养老困境。针对老年人意外伤害多发问题，提出基于环境布设的老年人异常行为检测系统设计，监测老年人的安全状态和构建健康的住宅环境。通过在室内布设传感器获取老人的行为数据，建立基于支持向量机算法的跌倒检测模型，实现对老人异常行为的检测和步态参数获取。研究结果表明，该系统的异常行为识别准确率达到95.71%,虚警率和漏检率分别为2.38%和0。此外，系统收集的老年人步态参数能够辅助用于跌倒预测和健康状况评估。本系统有助于实现老年人室内环境下意外伤害的早发现、早预防，为家居环境的适老化改造提供了新的解决方案，在智能健康住宅领域具有较好的应用前景。更多还原

【Abstract】 With the intensification of the aging process, the living environment based on intelligent health design is expected to solve the social pension dilemma. In view of the frequent occurrence of accidental injuries in the elderly, the design of the elderly abnormal behavior detection system based on environment layout was proposed to monitor the safety status of the elderly and build a healthy residential environment. By installing sensors in the room to obtain the behavior data of the elderly, a fall detection model based on Support Vector Machine algorithm was established to detect the abnormal behavior of the elderly and obtain gait parameters. The research results show that the accuracy of abnormal behavior recognition of the system is 95.71%, and the false alarm rate and missed detection rate are 2.38% and 0, respectively. In addition, the gait parameters of the elderly collected by the system can be used for fall prediction and health assessment. This system is helpful to realize the early detection and prevention of accidental injuries in the indoor environment of the elderly, which provides a new solution for the age-appropriate renovation of the home environment for older adults and has a good application prospect in the field of intelligent health housing.更多还原