reesefrancis
- 3
- 0
Hi, I have to find the best approach for tackling a problem for trying to recognize physical movements - with an iPhone in a pocket - like walking, stopping, turning left/right, sitting.
The ultimate goal is to recognize urban street behaviour, mostly regarding traffic lights: is it possible to tell when a pedestrian stops at a red light and then goes across the road on a green light? Or the data from an accelerometer won't be different when walking in a park etc.
I was thinking on just heuristically find the data corresponding to each action, then to check the incoming values against this data (with a threshold) and see what's happening. That's a very rough approach, of course, but unfortunately I don't have time to set up Support Vector Machine method for recognizing my patterns.
Here's what I got:
Walking: Do an fft on the gravity direction signal. Measure its frequency response for walking at different speeds and then set a simple threshold.
Stopping: if the average power i.e. total energy in the signal over the last few seconds drops below a certain threshold then you can say the user has stopped.
Turning left/right: not possible without a gyroscope.
Sitting: with no idea here - except for collecting data when sitting up from standing up and viceversa.
Stair climbing: basically the data I get when I climb stairs isn't different from the one I get when walking. Or is it there some way to tell the difference?
The ultimate goal is to recognize urban street behaviour, mostly regarding traffic lights: is it possible to tell when a pedestrian stops at a red light and then goes across the road on a green light? Or the data from an accelerometer won't be different when walking in a park etc.
I was thinking on just heuristically find the data corresponding to each action, then to check the incoming values against this data (with a threshold) and see what's happening. That's a very rough approach, of course, but unfortunately I don't have time to set up Support Vector Machine method for recognizing my patterns.
Here's what I got:
Walking: Do an fft on the gravity direction signal. Measure its frequency response for walking at different speeds and then set a simple threshold.
Stopping: if the average power i.e. total energy in the signal over the last few seconds drops below a certain threshold then you can say the user has stopped.
Turning left/right: not possible without a gyroscope.
Sitting: with no idea here - except for collecting data when sitting up from standing up and viceversa.
Stair climbing: basically the data I get when I climb stairs isn't different from the one I get when walking. Or is it there some way to tell the difference?