Fall detection apparatus and method
Provided are a fall detection apparatus and method. The fall detection method may include measuring first to third accelerations corresponding to a user, measuring first to third angular velocities corresponding to the user, determining an acceleration value for the user based on at least one of the first to third accelerations, determining an angular velocity value for the user based on at least one of the first and second angular velocities, determining an angle value for the user based on at least one of the first to third accelerations, calculating a triangle feature using the acceleration value, and detecting a fall of the user based on at least one of the acceleration value, the angular velocity value, the angle value, and the triangle feature.
1. A fall detection method, comprising:
measuring first to third accelerations corresponding to a user;
measuring first to third angular velocities corresponding to the user;
determining an acceleration value for the user based on at least one of the first to third accelerations;
determining an angular velocity value for the user based on at least one of the first and second angular velocities;
determining an angle value for the user based on at least one of the first to third accelerations;
calculating a triangle feature using the acceleration value; and
detecting a fall of the user based on at least one of the acceleration value, the angular velocity value, the angle value, and the triangle feature.
2. The fall detection method of claim 1 , wherein the first to third accelerations are an x-axis acceleration, a y-axis acceleration, and a z-axis acceleration, respectively, and
the first and second angular velocities are a pitch angular velocity and a roll angular velocity, respectively.
3. The fall detection method of claim 1 , wherein the acceleration value is calculated through a square root for a sum of squares of the first to third accelerations, respectively.
4. The fall detection method of claim 3 , wherein the acceleration value is calculated by the following [Equation 1]:
ACC SVM =√{square root over (ACC x 2 +ACC y 2 +ACC z 2 )} [Equation 1]
where ACC SVM is the acceleration value, ACC x is the first acceleration, ACC y is the second acceleration, and ACC z is the third acceleration.
5. The fall detection method of claim 4 , wherein the detecting of the fall further includes detecting the fall of the user by further considering a comparison result of the angular velocity value and a second reference value.
6. The fall detection method of claim 1 , wherein the angular velocity value is calculated through a square root for a sum of squares of the first and second angular velocities, respectively.
7. The fall detection method of claim 6 , wherein the angular velocity value is calculated by the following [Equation 2]:
ω SVM =√{square root over (ωPitch 2 +ωRoll 2 )} [Equation 2]
where ω SVM is the angular velocity value, ω Pitch is the first angular velocity, and ω Roll is the second angular velocity.
8. The fall detection method of claim 7 , wherein the detecting of the fall further includes detecting the fall of the user by further considering a comparison result of the triangle feature and a third reference value.
9. The fall detection method of claim 1 , wherein the detecting of the fall further includes detecting the fall of the user by considering a comparison result of the acceleration value and a first reference value.
10. The fall detection method of claim 9 , wherein the detecting of the fall further includes detecting the fall of the user when the acceleration value is smaller than the first reference value, the angular velocity value is greater than the second reference value, and the triangle feature is greater than the third reference value.
11. The fall detection method of claim 10 , wherein the first reference value is 0.9 g, the second reference value is 47.3°/s, and the third reference value is 0.19.
12. The fall detection method of claim 10 , further comprising:
determining the first reference value to the third reference value based on information on the user.
13. The fall detection method of claim 12 , wherein the information on the user includes at least one of an age, a gender, a body dimension, and a weight.
14. The fall detection method of claim 13 , further comprising:
outputting information corresponding to the detected fall from a sensing unit using the information on the user as an input value.
15. The fall detection method of claim 14 , wherein the sensing unit generates the input value to correspond to the age, the gender, the body dimension, and the weight, respectively, and
differently generates output information according to a generated comparison value based on output reference values allocated to each of the input values.
16. The fall detection method of claim 15 , further comprising:
outputting the output information through at least one of a sound, an image, data, a signal, or light.
17. A fall detection apparatus, comprising:
an acceleration measurement unit that measures first to third accelerations corresponding to a user;
an angular velocity measurement unit that measures first to third angular velocities corresponding to the user;
an acceleration value determination unit that determines an acceleration value for the user based on at least one of the first to third accelerations;
an angular velocity value determination unit that determines an angular velocity value for the user based on at least one of the first and second angular velocities;
an angle value determination unit that determines an angle value for the user based on at least one of the first to third accelerations;
a triangle feature calculation unit that calculates a triangle feature using the acceleration value; and
a fall detection unit that detects a fall of the user based on at least one of the acceleration value, the angular velocity value, the angle value, and the triangle feature.