Development of database on daily living activities for realistic biomechanical simulation

This paper describes the development of the fall database for a biomechanical simulation. First, data of children's daily activities were collected at a sensor home, which is a mock daily living space. The sensor home comprises a video-surveillance system embedded into a daily-living environment and a wearable acceleration-gyro sensor. Then, falls were detected from sensor data using a fall detection algorithm developed by authors, and videos of detected falls were extracted from long-time recorded video. The extracted videos were used for fall motion analysis. A new computer vision (CV) algorithm was developed to automate fall motion analysis. Using the developed CV algorithm, fall motion data were accumulated into a database. The developed database allows a user to perform conditional searches of fall data by inputting search conditions, such as a child's attributes and fall situation. Finally, a biomechanical simulation of falls was conducted with initial conditions set using the database.