Heart rate variability and emotion analysis method and system based on rPPG and rBCG fusion

By fusing rPPG and rBCG signals, HRV features are extracted from facial videos, solving the problems of complexity in traditional HRV acquisition methods and susceptibility to interference with non-contact signals, thus achieving high-precision emotion state recognition and daily monitoring.

CN121834401APending Publication Date: 2026-04-10NORTHEASTERN UNIV CHINA
View PDF 0 Cites 4 Cited by

Patent Information

Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
NORTHEASTERN UNIV CHINA
Filing Date
2025-12-22
Publication Date
2026-04-10

AI Technical Summary

Technical Problem

Traditional methods of acquiring HRV rely on contact sensors, which makes data collection complex and unsuitable for daily emotion monitoring. Non-contact signals are easily affected by environmental and individual factors, leading to a decrease in the stability and accuracy of heart rate detection, which in turn affects the accuracy of emotion state analysis.

Method used

By fusing remote photoplethysmography (rPPG) and remote cardiac impaction (rBCG) signals, signals are extracted from facial videos using a multi-channel detection and tracking method. These signals are then preprocessed, aligned, and fused to generate an HRV feature set, which is then accurately analyzed using HRV analysis and emotion classification models.

Benefits of technology

It improves the robustness of heart rate detection and the accuracy of emotion state recognition, enhances the convenience and comfort of non-contact measurement, can stably identify emotions in complex environments, and significantly improves the accuracy and reliability of emotion recognition.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN121834401A_ABST
    Figure CN121834401A_ABST
Patent Text Reader

Abstract

The invention provides a heart rate variability and emotion analysis method and system based on rPPG and rBCG fusion, and relates to the technical field of biomedical signal processing. The method specifically comprises the following steps: extracting an original rPPG signal and an original rBCG signal from a face video of a subject, and preprocessing the original rPPG signal and the original rBCG signal to obtain a basic rPPG signal and a basic rBCG signal; a fused heart beat signal sequence is generated through alignment and fusion; heart beat peak detection is carried out on the fused heart beat signal sequence, a heart beat interval sequence is generated, time domain indexes and frequency domain indexes of the heart beat interval sequence are calculated, and an HRV feature set is obtained; and based on the fused heart beat signal sequence and the HRV feature set, performing HRV prediction and emotion classification by using the trained HRV analysis and emotion classification model, and generating an HRV prediction result and an emotion classification result of the subject. According to the invention, the robustness of heart beat detection and the accuracy of emotional state recognition can be improved under a non-contact measurement condition.
Need to check novelty before this filing date? Find Prior Art

Citation Information

Cited By

  • Methods, devices, equipment, and media for monitoring high physiological stress based on rPPG

    CN122136004A

  • Methods, devices, equipment, and media for monitoring high physiological stress based on rPPG

    CN122136004B

  • Layered progressive ppg signal generation method for data augmentation

    CN122271985A

  • Layered progressive ppg signal generation method for data augmentation

    CN122271985B