T O P I C R E V I E W |
freefeynman123 |
Posted - 07/05/2016 : 1:00:28 PM I'm trying to write a script which fits linear combination of 4 Gaussian lines to multiple spectra (1 column with x rows and 51 with data) with suggested, but not fixed positions (something like Multiple Peak Fit, but used to many spectra). I tried a code like this:
loop (num, 2, 52) { nlbegin iy:=(1,num) func:=gauss + gauss + gauss + gauss; }
but I don't know how to assign suggested peak positions to different Gaussian lines. I would appreciate any help. |
3 L A T E S T R E P L I E S (Newest First) |
Hideo Fujii |
Posted - 07/07/2016 : 11:01:56 AM Hi freefeynman123,
Finding hidden peaks requires appropriate treatment to succeed as described at: http://www.originlab.com/doc/Origin-Help/FindPeak-PA
I have tried the Peak Analyzer (Goal=FitPeaks(Pro)) with four initial positions of 4 peaks manually at the bumps, and got the following result:
Maybe you can first try this method, then you can consider the batch processing with Peak Analyzer?
Hope this info helps.
--Hideo Fujii Originlab
|
freefeynman123 |
Posted - 07/07/2016 : 04:55:04 AM Thank you for the response, however the result is different from what I desired. I would like to have a fit like this:
However what I get looks like this:
there are no plots of single Gaussian peaks and I have no information about position, intensity etc. and it just fits to the one spectra, not the whole data.
I would really appreciate further help. |
Hideo Fujii |
Posted - 07/05/2016 : 2:04:37 PM Hi freefeynman123,
Here is a sample code of "Fit Multiple Peaks with Replica" with Gauss function with 3 replicas:
http://www.originlab.com/doc/LabTalk/examples/Curve-Fitting#Fit_Multiple_Peaks_with_Replica
Please try this sample.
--Hideo Fujii OriginLab |