[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
Re: DSP - pattern search
|
From: |
dynamphorous |
|
Subject: |
Re: DSP - pattern search |
|
Date: |
Tue, 10 May 2011 09:19:46 -0700 (PDT) |
So you are saying that it changes enough that you cant just use the xcorr
function to meaningfully look for the cross correlation of the dataset and
the underlying shape? It is pretty good at picking out shapes in noise.
The other thing that immediately comes to mind would be a artificial neural
network trained with backpropagation. Those are quite good at identifying
patterns, even corrupted ones surrounded by noise.
--
View this message in context:
http://octave.1599824.n4.nabble.com/DSP-pattern-search-tp3512334p3512376.html
Sent from the Octave - General mailing list archive at Nabble.com.