유헬스 환경을 위한 비접촉 무구속 호흡 신호 측정 시스템
한태화1), 이영호2), 강운구2), 서영준1), 이경진1), 황성철1), 박래웅*1)
1)아주대학교 의과대학 의료정보학과
Abstract : This study aims to reveal the practicability of unobtrusive remote respiration detection method using image analysis. Five subjects were enrolled in this experiment. The minute movements of subject's trunk by respiration are captured by a infrared-pass-filter equipped camera while infrared-ray illuminates the subject. The signals were converted into image difference value between the current frame and the previous frame. The frequencies of signal are analyzed by Fourier transform. The detected respiration rates were compared with the real respiration rates. Each of the experiments were repeatedly conducted with different respiration rate, such as normal, slow, and fast phase. The overall accuracy of the unobtrusive respiration detection method using image difference is ±93.3%. Statistically, there was no difference between real respiration rate and the respiration rate detected by image analysis. keyword : contact-free monitoring, unobtrusive monitoring, respiration rate, image analysis
, , , ,
투고내용을 삭제하시려면 "삭제"버튼을 누르시고 다시 접수하여 주시기 바랍니다.
Copyright (c) 2002 by The Korean Society of Medical Informatics