Fall detection system and method
A fall detection system includes a radar that generates emitting radio waves and receives reflected radio waves from a person under detection, a data generator that generates a point cloud according to the reflected radio waves, an area determining device that determines a sub-area of a detecting area in which the person under detection lies, and a classifier that determines whether the person under detection falls according to the point cloud. The classifier adaptively processes the point cloud with different methods according to sub-areas as determined by the area determining device respectively to determine whether the person under detection falls.
1. A fall detection system, comprising:
a radar that generates emitting radio waves and receives reflected radio waves from a person under detection;
a data generator that generates a point cloud according to the reflected radio waves;
an area determining device that determines a sub-area of a detecting area in which the person under detection lies; and
a classifier that determines whether the person under detection falls according to the point cloud;
wherein the classifier adaptively processes the point cloud representing the person under detection with different methods according to sub-areas where the person under detection lies as determined by the area determining device respectively to determine whether the person under detection falls.
2. The system of claim 1 , wherein the radar comprises a millimeter wave radar.
3. The system of claim 1 , wherein the detecting area is divided into the following sub-areas:
an ordinary zone that is an area inside a coverage area of the radar or a field of view line;
a dead zone that is an area to which the radar is close and therefore cannot properly receive the reflected radio waves; and
a peripheral zone that is an area outside the coverage area of the radar or the field of view line.
4. The system of claim 3 , wherein radio wave intensity in the ordinary zone is substantially greater than radio wave intensity in the dead zone or the peripheral zone.
5. The system of claim 1 , wherein the classifier comprises a neural network trained to extract features from the point cloud in order to determine whether the person under detection falls.
6. The system of claim 3 , wherein the classifier determines a size and a rate of height change of the point cloud to determine whether the person under detection falls, when the person under detection is determined to lie in the ordinary zone.
7. The system of claim 6 , wherein the person under detection is decided to be falling when the size of the point cloud is less than a predetermined first threshold and the rate of height change of the point cloud is greater than a predetermined second threshold.
8. The system of claim 3 , wherein the classifier determines whether the point cloud moves in a predefined period, when the person under detection is determined not to lie in the ordinary zone.
9. The system of claim 8 , wherein the person under detection is decided to be falling when moving distance of the point cloud in the predetermined period is less than a predetermined threshold.
10. A fall detection method, comprising:
generating emitting radio waves and receiving reflected radio waves from a person under detection;
generating a point cloud according to the reflected radio waves;
determining a sub-area of a detecting area in which the person under detection lies; and
determining whether the person under detection falls according to the point cloud;
wherein the point cloud representing the person under detection is adaptively processed with different methods according to sub-areas where the person under detection lies as determined respectively to determine whether the person under detection falls.
11. The method of claim 10 , wherein the emitted radio waves are generated by a radar.
12. The method of claim 11 , wherein the detecting area is divided into the following sub-areas:
an ordinary zone that is an area inside a coverage area of the radar or a field of view line;
a dead zone that is an area to which the radar is close and therefore cannot properly receive the reflected radio waves; and
a peripheral zone that is an area outside the coverage area of the radar or the field of view line.
13. The method of claim 12 , wherein radio wave intensity in the ordinary zone is substantially greater than radio wave intensity in the dead zone or the peripheral zone.
14. The method of claim 10 , wherein the point cloud is processed by a neural network trained to extract features from the point cloud in order to determine whether the person under detection falls.
15. The method of claim 12 , wherein a size and a rate of height change of the point cloud is determined to decide whether the person under detection falls, when the person under detection is determined to lie in the ordinary zone.
16. The method of claim 15 , wherein the person under detection is decided to be falling when the size of the point cloud is less than a predetermined first threshold and the rate of height change of the point cloud is greater than a predetermined second threshold.
17. The method of claim 12 , wherein the point cloud is determined whether moves in a predefined period, when the person under detection is determined not to lie in the ordinary zone.
18. The method of claim 17 , wherein the person under detection is decided to be falling when moving distance of the point cloud in the predetermined period is less than a predetermined threshold.