http://chineseinput.net/에서 pinyin(병음)방식으로 중국어를 변환할 수 있습니다.
변환된 중국어를 복사하여 사용하시면 됩니다.
CCTV 와 레이더 센서를 활용한 실내에서의 낙상 사고 감지 방법
김유정(YuJeong Kim),김재철(JaeCheol Kim),김효정(HyoJeong Kim),이종혁(JoungHyeok Lee),정준호(Junho Jeong) 대한전자공학회 2023 대한전자공학회 학술대회 Vol.2023 No.6
Research is currently underway on real-time fall detection in indoor environments using pose estimation models applied to CCTV. While action recognition models exhibit high accuracy in frontal-view, they suffer from decreased accuracy when applied to CCTV footage setting at the edges of the ceiling where distortion occurs. To address this issue, this paper proposes a method for fall accident detection by combining CCTV with a radar sensor. The proposed pose estimation model combines YOLO, AlphaPose, and ST-GCN to measure seven actions (standing, walking, sitting, stand up, lying down, sitting, fall down). The data values from the radar sensor are labeled and trained using logistic regression to determine fall accidents. Additionally, weights are applied to the measurements of the seven actions.