메뉴 건너뛰기
소속 기관 / 학교 인증
인증하면 논문, 학술자료 등을  무료로 열람할 수 있어요.
한국대학교, 누리자동차, 시립도서관 등 나의 기관을 확인해보세요
(국내 대학 90% 이상 구독 중)
고객센터 ENG
주제분류

논문 기본 정보

저자정보
(광운대학교) (광운대학교)
저널정보
대한전자공학회 대한전자공학회 학술대회 2025년도 대한전자공학회 하계학술대회 논문집
오류 신고하기

피인용 0

검색

    초록·키워드

    Remote photoplethysmography (rPPG) enables non-contact heart rate (HR) estimation using facial videos, offering advantages in terms of convenience and hygiene. In this study, we present a signal processing pipeline for rPPG extraction based on Singular Value Decomposition (SVD). Facial regions are detected using Mediapipe FaceMesh and masked to exclude eye and mouth areas. The RGB time- series signal is constructed by spatial averaging of masked regions and then projected using a principal component–based method. Specifically, the dominant direction of illumination variation is removed using a projection matrix derived from the first left-singular vector of the RGB signal matrix. The second column of the projected signal is selected as the rPPG signal, followed by detrending and bandpass filtering. Our method is evaluated on the PURE and UBFC-rPPG datasets and demonstrates competitive HR estimation performance without requiring any supervised training. This highlights the potential of lightweight, interpretable signal processing methods for physiological monitoring in real-world settings.

    최근 본 자료 전체보기

      UCI(KEPA) : I410-151-25-02-093761560