I am using a tri-axis accelerometer to capture the movements of the hand. While coming through the existing solutions, I found that FastDTW is a suitable one for identifying the patterns of a time series data set. I came across the following libraries.
https://code.google.com/p/fastdtw/ https://github.com/cscotta/fastdtw/tree/9322f2a74c96f9c38f253c702a57dd23a16cf777
I'm new to this and I want to know, whether there are any useful examples that i could use to start playing with that. What i want to do is to give the Accelerometer singals as input and identify the pattern so that it would represent a particular movement. I would highly appreciate if you could provide some useful code example for this.
Thank You.
yes, the DTW algorithm should work, however getting consistent readings is the hard part.