基于嵌入式与表情识别的人体生理参数监测系统设计

Design of human physiological parameter monitoring system based on embedded and expression recognition

摘要:

为满足对人体健康信息实时且多维度监测，提高监护人员工作效率，设计了一套人体生理参数和表情识别的健康监测系统。通过传感器、摄像头等设备，实时收集人体生理指标和视频资料，核心控制层和服务器进行集中分析处理，监测结果通过PC端界面展示，帮助监护人员实时关注被监护人身体状况和情绪变化，及时采取干预措施。实验结果表明该系统对人体血氧饱和度监测平均误差在0.6%以内；对人体心率监测平均误差在0.3 bpm以内；对人体体温监测平均误差在0.13℃以内；对人脸表情识别的准确率为96.62%，其具有监测人体多维度健康信息的能力，可以解决实时监测人体生理参数和情绪变化的问题，有效提高监护人员的工作效率。

Abstract:

In order to meet the needs of real-time and multi-dimensional monitoring of human health information and improve the work efficiency of guardians, a health monitoring system for human physiological parameters and expression recognition is designed. Through sensors, cameras and other equipment, human physiological indicators and video data are collected in real time. The core control layer and server perform centralized analysis and processing, and the monitoring results are displayed through the PC interface. It helps guardians pay real-time attention to the physical condition and emotional changes of the ward and take timely intervention measures. The experimental results show that the average error of the system in monitoring human blood oxygen saturation is within 0.6%; the average error in monitoring human heart rate is within 0.3 bpm; the average error in monitoring human body temperature is within 0.13°C; the accuracy of facial expression recognition is 96.62%, it has the ability to monitor multi-dimensional health information of the human body, can solve the problem of real-time monitoring of human physiological parameters and emotional changes, and effectively improve the work efficiency of monitoring personnel.