APPARATUS AND METHOD FOR REAL-TIME RECOGNITION OF ACTIVITY AND POSTURE BASED ON COMBINED SENSOR DATA
The present invention relates to a device for recognizing the posture and activity of a user on a real-time based on hybrid sensor data and a method thereof. The device uses the hybrid sensor data of a fixed sensor and a non-fixed sensor to recognize the posture and activity of the user. The device...
Saved in:
Main Authors | , , |
---|---|
Format | Patent |
Language | English Korean |
Published |
14.01.2016
|
Subjects | |
Online Access | Get full text |
Cover
Loading…
Summary: | The present invention relates to a device for recognizing the posture and activity of a user on a real-time based on hybrid sensor data and a method thereof. The device uses the hybrid sensor data of a fixed sensor and a non-fixed sensor to recognize the posture and activity of the user. The device can be used to recognize the posture and activity of the user in various environments more accurately. The device includes a windowing part for achieving exercise data; a recognition model determining part for processing the achieved exercise data to the hybrid sensor data and verifying the extracted feature information to determine the recognition model; and a posture and activity classification part for outputting the classification result for the posture and activity of the user in real time.
고정식 센서 및 비고정식 센서의 혼합 센서 데이터를 이용하여 사용자의 행위 및 자세를 인지하는 혼합 센서 데이터 기반 실시간 행위인지 및 자세인지 장치 및 그 방법을 제공하여 다양한 환경에서의 사용자의 행위 및 자세를 보다 정확하게 인지할 수 있다. |
---|---|
Bibliography: | Application Number: KR20140174075 |