상세정보
학술지
Fall Recognition System to Determine the Point of No Return in Real?Time
- 발행일
- 202109
- 출처
- Applied Sciences, v.11 no.18, pp.1-12
- ISSN
- 2076-3417
- 출판사
- MDPI
- 협약과제
- 20HS4100, 지능정보 및 메타 소재‧구조물 기술 기반의 노약자 보행지원을 통한 낙상예측‧방지 소프트 웨어러블 슈트 기술 개발, 손용기
- 초록
- In this study, we collected data on human falls, occurring in four directions while walking or standing, and developed a fall recognition system based on the center of mass (COM). Fall data were collected from a lower?륿ody motion data acquisition device comprising five inertial measurement unit sensors driven at 100 Hz and labeled based on the COM?릒orm. The data were learned to classify which stage of the fall a particular instance belongs to. It was confirmed that both the rep-resentative convolutional neural network learning model and the long short?릘erm memory learning model were performed within a time of 10 ms on the embedded platform (Jetson TX2) and the recognition rate exceeded 94%. Accordingly, it is possible to verify the progress of the fall during the unbalanced and falling steps, which are classified by subdividing the critical step in which the real?릘ime fall proceeds with the output of the fall recognition model every 10 ms. In addition, it was confirmed that a real?릘ime fall can be judged by specifying the point of no return (PONR) near the point of entry of the falling down stage.
- KSP 제안 키워드
- Convolution neural network(CNN), Data Acquisition(DAQ), Inertial Measurement Unit(IMU), Learning model, Motion Data, Neural network learning, Recognition System, Recognition model, Recognition rate, center of mass, data acquisition device
본 저작물은 크리에이티브 커먼즈 저작자 표시 (CC BY) 조건에 따라 이용할 수 있습니다.
