Presently, numerous applications related to ubiquitous healthcare have developed with sensor and body sensor network technologies. In the same approach, this study proposed a 3D Falling Reconstruction system to detect the occurrence of accidental falling using a number of accelerometers, gyroscopes, and wireless modules. Upon the occurrence of falling, the system uses kinematic formulas to calculate the motions of human skeleton, and reconstructs body posture, falling process, and impacts of major body parts in a 3D scene, which provide more information for medical personnel to make accurate judgments. The main architecture of the system consists of two parts, the detection of accidental falling and the reconstruction of elderly body posture. In the first part, because of the clustering property of human motion in accordance with personal habits, the system detects the occurrence of accidental falling by utilizing subtractive clustering algorithm with motion trajectory data from accelerometers, and the shift degree of barycenter. The work in the second part includes establishing data structures of human skeleton, and reconstructing the complete falling process.
All Science Journal Classification (ASJC) codes
- Atomic and Molecular Physics, and Optics
- Electrical and Electronic Engineering