摘要
在使用毫米波雷达进行室内人员信息检测时,其信号处理阶段采用的静态杂波滤除算法有效地滤除了检测区域中包括墙壁、地面、桌椅等在内的静止目标,实现了对运动人员的检测,但同时会导致静止人员被漏检.为此提出按照径向速度把点云数据划分为动态数据和静态数据,先剔除动态数据,然后累积剩余的静态数据.在达到指定的累积帧数时,进行密度聚类,以簇的数量作为人员的数量,簇的中心坐标作为人员的位置.通过实验,验证了所提出方法的有效性,在室内办公场景下,人员数量统计平均绝对误差为0.81,人员位置估计均方根误差为0.1 m.
When millimeter wave radar is used for indoor occupancy information detection,the static clutter filtering algorithm adopted in the signal processing stage can effectively filter out the static targets in the detection area,such as walls,ground,desks and chairs,and the moving people are detected,but at the same time,the static people will be missed.To solve this problem,the point cloud data are divided into dynamic data and static data according to the radial velocity,firstly the dynamic data are removed,and then the remaining static data are accumulated.When the specified cumulative frame number is reached,density clustering is carried out,and the number of clusters is taken as the number of people,and the center coordinate of the cluster is taken as the location of people.The effectiveness of the proposed method is verified by experiments.In the indoor office scenario,the mean absolute error of occupants counting is 0.81,and the root mean square error of the location estimation of people is 0.1 m.
作者
赵亮
李论
ZHAO Liang;LI Lun(School of Control Science and Engineering,Dalian University of Technology,Dalian 116024,China)
出处
《大连理工大学学报》
CAS
CSCD
北大核心
2024年第4期418-425,共8页
Journal of Dalian University of Technology
基金
国家自然科学基金资助项目(52178066)。
关键词
人员数量统计
室内人员信息检测
毫米波雷达
密度聚类
occupants counting
indoor occupancy information detection
millimeter wave radar
density clustering
