Determining body postures and activities
A method and wireless sensor device for determining body postures and activities. In one aspect, a method includes receiving sensor data. The method also includes detecting and classifying a body transition of a body based on the sensor data. The method also includes detecting if there is activity of the body based on the sensor data. If there is activity, the method also includes classifying the activity. If there is no activity, the method also includes classifying a rest position of the body based on the sensor data and based on a previous body transition.
1. A computer-implemented method for determining body postures and activities, the method comprising:
obtaining sensor data using a body sensor device, wherein the body sensor device comprises a tri-axial accelerometer and the sensor data includes tri-axial accelerometer data;
obtaining calibrated sensor data by applying rotation of accelerometer axes to the tri-axial accelerometer data to get rotated accelerations in three calibrated axes;
calculating a transition signal from the calibrated sensor data as acceleration on Y-axis minus acceleration on Z-axis to:
detect a transition when a first body position transitions to a second body position, and
classify the transition as a particular body transition if the transition signal exceeds a transition threshold; and
determining if there is activity of the body based on the calibrated sensor data, classifying an activity if there is activity, and classifying a rest position of the body if there is no activity wherein the rest position of the body is classified based on the sensor data and based on a previous body transition.
2. The method of claim 1 , further comprising:
calibrating the sensor data such that axes of the body sensor device matches axes of the body.
3. The method of claim 1 , wherein the body transition is any of sitting-to-standing, standing-to-sitting, and bending over.
4. The method of claim 1 , wherein the determining if there is activity of the body further comprises:
determining a signal magnitude area of accelerometer signals of the sensor data; and
determining if the signal magnitude area is greater than an activity threshold, wherein the activity is detected if the signal magnitude area is greater than the activity threshold.
5. The method of claim 1 , wherein the activity is any of walking and running.
6. The method of claim 1 , wherein the classifying of the rest position of the body further comprises:
calculating a vertical angle; and
classifying the rest position based on the vertical angle and the previous body transition.
7. The method of claim 1 , wherein the rest position is any of sitting, standing, and lying down.
8. A body sensor device, comprising:
a sensor for obtaining sensor data, wherein the sensor comprises a tri-axial accelerometer and the sensor data includes tri-axial accelerometer data;
a processor coupled to the sensor; and
a memory device coupled to the processor, wherein the body sensor device includes an application that, when executed by the processor, causes the processor to perform operations comprising:
obtaining calibrated sensor data by applying rotation of accelerometer axes to the tri-axial accelerometer data to get rotated accelerations in three calibrated axes;
calculating a transition signal from the calibrated sensor data as acceleration on Y-axis minus acceleration on Z-axis to:
detect a transition when a first body position transitions to a second body position, and
classify the transition as a particular body transition if the transition signal exceeds a transition threshold; and
determining if there is activity of the body based on the calibrated sensor data, classifying an activity if there is activity, and classifying a rest position of the body if there is no activity wherein the rest position of the body is classified based on the sensor data and based on a previous body transition.
9. The body sensor device of claim 8 , wherein the processor further performs operations comprising:
calibrating the sensor data such that axes of the body sensor device matches axes of the body.
10. The body sensor device of claim 8 , wherein the body transition is any of sitting-to-standing, standing-to-sitting, and bending over.
11. The body sensor device of claim 8 , wherein the determining if there is activity of the body further comprises:
determining a signal magnitude area of accelerometer signals of the sensor data; and
determining if the signal magnitude area is greater than an activity threshold, wherein the activity is detected if the signal magnitude area is greater than the activity threshold.
12. The body sensor device of claim 8 , wherein the activity is any of walking and running.
13. The body sensor device of claim 8 , wherein the classifying of the rest position of the body further comprises:
calculating a vertical angle; and
classifying the rest position based on the vertical angle and the previous body transition.
14. The body sensor device of claim 8 , wherein the rest position is sitting.
15. The body sensor device of claim 8 , wherein the rest position is standing.
16. The body sensor device of claim 8 , wherein the rest position is lying down.