嵌入式无接触视频心率检测方法  

Contactless Videosequence Heart Rate Detection by Embedded System

作　　者：周秦武[1] 隋芳芳[1] 白平[1] 宋亚楠[1] 尹小龙[1] 

机构地区：[1]西安交通大学生物医学信息工程教育部重点实验室,西安710049

出　　处：《西安交通大学学报》2013年第12期55-60,共6页Journal of Xi'an Jiaotong University

摘　　要：为了解决方便快捷的心率检测问题,实现心血管疾病的常态监护,提出了一种将嵌入式技术与视频检测心率技术结合起来的心率检测方法。该方法利用人体心血管活动会使血管容积周期性变化,从而影响体表对光照的吸收和反射,通过捕捉体表反射光变化分析心血管活动的时序。采用嵌入式便携监护终端和图像传感器完成人体视频数据的采集,通过对视频数据的局部区域进行三原色盲源分离和独立变量功率谱密度分析得到受试者心率参数。实验结果表明,面部区域视频计算的受试者心率与心电图机结果一致,测量误差小于2.8%,检测灵敏度为1次/min,并且计算结果对光照等外界环境不敏感,表明本算法具有良好的可靠性和稳定性。For easily and quickly detecting heart rate and achieving norma｜ care of cardiovascular disease, a contactless heart rate detection scheme combining videosequence detection with embedded system is presented. The cardiovascular activity usually results in periodically changed vascular volume in each cycle, and affects the absorption and reflection of light on body surface. Capturing the changed reflected light from the superficial body, the cardiovascular activity timing can be analyzed. Human videosequence acquisition is completed by the embedded portable terminal with CMOS image sensor. Then the face detection, the blind source separation and the independent variables power spectral density analysis are conducted to evaluate heart rate parameters. The heart rate from our scheme is almost equal to that from electrocardiogram. The measurement error gets less than 2.8~ and sensitivity achieves 1 time/rain. Especially, the result is insensitive to light.

关 键 词：心率 视频图像 嵌入式系统 盲源分离 功率谱密度 

分 类 号：TN911.73[电子电信—通信与信息系统]