Prev: synchronize two three phase voltage supplies
Next: TOMLAB optimization
From: ImageAnalyst on 30 Nov 2009 06:40
What about doing regular "outlier detection"?
http://www.google.com/search?hl=en&source=hp&q=outlier+detection&aq=1&oq=outlier+d&aqi=g10

Just filter the data with a running average (filter() or conv()) and
std() and then scan it to see if the data is farther away from the
mean than 2 stddev's or so.

By the way, 3000 data points is a microscopic amount of data these
days. Even a garden variety digital camera can produce 10 million
data points (pixels).
First  |  Prev  | 
Pages: 1 2
Prev: synchronize two three phase voltage supplies
Next: TOMLAB optimization