摘要
提出复杂环境下基于特征融合的日常动作、突发异常(摔倒)行为检测方法.利用人的姿态、姿态变化速率特征、人的位置变化特征表征人的运动状态,通过合成简单的姿态事件并结合特征来表达具有复杂时空关系的运动事件.该方法计算复杂度小,对目标大小的变化具有较好的鲁棒性,在智能交互、服务机器人自主服务系统中具有实用价值.
A novel method was proposed to detect human daily activities and fall behavior based on features fusion. Actions were described by a set of postures and features of postures change velocity and position changes, then simple posttLres events and features were combined to express complex human activities events. This method can robustly detect human actions, thus can be used for intelligence interactive and service robot autonomous server system.
出处
《山东大学学报（工学版）》
CAS
北大核心
2009年第5期43-47,共5页
Journal of Shandong University（Engineering Science）
基金
国家高技术研究发展计划(863计划)家庭服务机器人重点基金资助项目(2006AA040206)
留学回国人员科研启动基金资助项目(教外司留[2007]24号)
关键词
智能监护
运动检测
人体姿态识别
行为理解
智能空间
Intelligent surveillance
motion detection
human postures recognition
behaviors understanding
intelligent space
