Scholarlink

The human heart rate serves as a critical metric of physiological condition and a significant indicator of an individual’s circulatory system. To measure heart rate, a non-intrusive facial recognition method has been proposed. The framework’s architecture comprises both software and hardware integration, enabling real-time heart rate measurement and validation. The system is developed with a series of steps including face detection, region of interest (ROI) extraction, green channel value extraction, signal processing, peak signal extraction, and heart rate calculation. The accuracy of the system’s heart rate measurements is validated against the heart rate results obtained from the IoT sensor, MAX30100, integrated with Arduino Mega. Testing involved 10 participants assessing the developed system in an indoor environment. Mean square error (MAE) and root mean squared error (RMSE) were calculated between the heart rate (BPM) measured by the system and the sensor to evaluate the difference and accuracy.

错误：搜索内容不能为空，请输入英文关键词
错误：关键词超出字数限制，请精简
高级检索

Heart Rate Measurement Using Face Recognition Techniques and Sensors

  • 会议论文

  • 发表日期：

    2025年

查看全文
  • Tan Hao Ze,
  • Sumendra Yogarayan,
  • Siti Fatimah Abdul Razak,
  • Mohd Fikri Azli Abdullah,
  • Afizan Azman

摘要

The human heart rate serves as a critical metric of physiological condition and a significant indicator of an individual’s circulatory system. To measure heart rate, a non-intrusive facial recognition method has been proposed. The framework’s architecture comprises both software and hardware integration, enabling real-time heart rate measurement and validation. The system is developed with a series of steps including face detection, region of interest (ROI) extraction, green channel value extraction, signal processing, peak signal extraction, and heart rate calculation. The accuracy of the system’s heart rate measurements is validated against the heart rate results obtained from the IoT sensor, MAX30100, integrated with Arduino Mega. Testing involved 10 participants assessing the developed system in an indoor environment. Mean square error (MAE) and root mean squared error (RMSE) were calculated between the heart rate (BPM) measured by the system and the sensor to evaluate the difference and accuracy.