04-26-2009 11:34 PM
Need to get the index of the minimize point, marked as red cycle in the attached word file.
It is a collected data from different sample devices, so the curve may change a bit each time. But the overall profile is the same, just 2 main peaks.
I tried to detect the Y difference to see when the difference values change from negative to positive.
But, there are noise as tiny peak when zoom in, shown in word file. Then tried to use median filter to make it smooth. (Maybe better method should be used.)
But there are still some bigger regional peak. Any suggestion?
Is there any labview function or better algorithm can make it. Many thanks for your help.
Solved! Go to Solution.
04-26-2009 11:40 PM
04-27-2009 12:51 AM
Thanks, it may solve the noise level small peak, but for wider regional peak, may I know how to remove.
Is there any way to detect it since the min point is quite sharp and odd then others.
Thanks.
04-27-2009 01:10 AM
How can u say the peak as wider regional peak,Do u want to remove only the minimum index data or u want to smoothen the entire data i tried with moving average filter find the attached vi.
04-27-2009 03:30 AM
As shown in the attached shot, there are still 2 more red cycles before the real one, but the difference of the 3 red cycles is:
The curve at the 3rd red cycle is not Smooth in math, that is its 1st order Derivative at the 3rd cyccle is not a Continuous function.
Is there any way to detect it?
04-27-2009 03:42 AM
A trick I often use is to sample the data with a much higher speed than needed. Then I apply some simple averaging methods or and median filter. After that I downsample the signal to a adequate rate by averaging x samples. This have been a saver for me more than once. How successful it is depends of noise character type. It just a hint as it may not be applicable in your application
04-27-2009 03:54 AM
04-27-2009 04:22 AM