基于多传感器的人体生理状态判别可视化技术  

Visualization discriminant technology for human body physiological status based on multi-sensor

作　　者：李思楠 赵海 LI Sinan;ZHAO Hai(Department of Physics and Electronic Engineering,Hebei Normal University for Nationalities,Chengde 067000,China;School of Computer Science and Engineering,Northeastern University,Shenyang 110000,China)

机构地区：[1]河北民族师范学院物理与电子工程系,河北承德067000 [2]东北大学计算机科学与工程学院,辽宁沈阳110000

出　　处：《传感器与微系统》2019年第12期25-28,32,共5页Transducer and Microsystem Technologies

基　　金：河北民族师范学院校级普通基金资助项目(PT2017008)

摘　　要：基于多传感器的人体生理状态判别可视化技术适用于当前的可穿戴体域网设备。该技术利用了3种可穿戴式人体生理信号传感器。从脉搏传感器中采集脉搏波信号,预处理后提取特征向量,采用支持向量机的方法,将人体生理状态分类为"普通状态"和"事件状态",对16名实验者都取得了90%以上的分类准确率。利用呼吸波传感器和体温传感器作为辅助判别方式,将这三种信号的分类结果采用二进制编码的方式进行数据融合,得出一种对人体生理状态的综合评价可视化结果。Visualization discriminant technology for human body physiological status based on multi-sensor is applicable to current wearable body area network devices.Three wearable human physiological signal sensors are utilized.The feature vector is extracted from the pre-processed pulse signal that acquired from pulse sensor,and the physiological status of the human body is classified into"normal status"and"event status"by the method of support vector machine.The classification accuracy rate is more than 90%for 16 experimenters.Respiratory wave sensor and body temperature sensor are also used as auxiliary methods.Results of visualization of comprehensive assessment of the physiological status of the human body is obtained according to the binary coding for data fusion of the three classification results.

关 键 词：多传感器 支持向量机 分类 二进制编码 数据融合 

分 类 号：TP212[自动化与计算机技术—检测技术与自动化装置] TP399[自动化与计算机技术—控制科学与工程]