摘要 |
Disclosed are a method and a system for establishing a human falling detection model. The method may comprise: collecting speed information in a human falling process in one or more human falling samples, the speed information comprising a resultant acceleration, speed, and time; analyzing features of the speed information, calculating and retrieving a lower acceleration threshold a0, a resultant acceleration sequence I0 within a fixed period of time, an upper acceleration threshold a1, a curve area threshold △S, and a lower acceleration time threshold △T, and establishing a falling detection mechanism. The system of the present invention takes full consideration of motion behavior characteristics of a human, increases a detection rate, and reduces an error detection rate. |