Peak detect

[Philippe Buhr]

How can I detect the few main peaks in a graph. The "peak detect" vi does
not sort the peaks detected, that's why it detects a lot of peaks instead of
the few main peaks.
What should I do ?

Thank you very much

Phil

[Tims]

Hi Philippe

Use the 'Peak Detect.vi' to detect the peaks and then use the 'Sort 1d Array' to sort the array of amplitudes from highest to
lowest.

To reduce the number of peaks detected to the few main peaks. Try increasing the peak threshold, so only peaks above a set level
are only detected.

Tim

Philippe Buhr wrote:

> How can I detect the few main peaks in a graph. The "peak detect" vi does
> not sort the peaks detected, that's why it detects a lot of peaks instead of
> the few main peaks.
> What should I do ?
>
> Thank you very much
>
> Phil

[John Wilson]

Hi Phil,
I'm currently working on that problem. Here at Sanders we do some Gas Chromatography work. We drag the damn GC out into the field in a 30' trailer all over the SouthEast to do testing. I've wrote a GC program once, but it sucks, and any way I wrote it many years ago, back in the old LabVIEW 3.0 days. The Peak Detection example you can get off the Nat'l Instruments FTP site sucks cause it either detects every frickin' peak in the world or you gotta do so much manipulation with thresholds and whatnot you'll have a stroke before you separate and integrate the peaks you want.
The VI I'm sending you is a work in progress, but it has some ideas that you can maybe expand and adapt for your needs. I just started writing the damned thing today (that's righ
t, I'm working on the 4th of July), so don't expect too much.
It detects a number of peaks specified by iMax. It detects the largest peak and eliminates them from the array, then detects the next largest, and eliminates that one, etc. It optionally reorders peaks at the end. I'm going to be working on this until I have a finished product, then I will post the VI on this forum.
Hope this helps.
John Wilson
Sanders Engineering & Analytical Services, Inc.
Mobile, Alabama
251-633-4120
johnwilson@pctechnician.net

[Mike Manzheliy]

Hi John,
I belief, Peak detection is very interesting problem.
And I designed an algorithm which contain probability function
Here it is
Mike Manzheliy
Automated Control Systems

John Wilson wrote:

> Hi Phil,
> I'm currently working on that problem. Here at Sanders we do some
> Gas Chromatography work. We drag the damn GC out into the field in a
> 30' trailer all over the SouthEast to do testing. I've wrote a GC
> program once, but it sucks, and any way I wrote it many years ago,
> back in the old LabVIEW 3.0 days. The Peak Detection example you can
> get off the Nat'l Instruments FTP site sucks cause it either detects
> every frickin' peak in the world or you gotta do so much manipulation
> with thresholds and whatnot you'll have a stroke bef
ore you separate
> and integrate the peaks you want.
> The VI I'm sending you is a work in progress, but it has some ideas
> that you can maybe expand and adapt for your needs. I just started
> writing the damned thing today (that's right, I'm working on the 4th
> of July), so don't expect too much.
> It detects a number of peaks specified by iMax. It detects the
> largest peak and eliminates them from the array, then detects the next
> largest, and eliminates that one, etc. It optionally reorders peaks
> at the end. I'm going to be working on this until I have a finished
> product, then I will post the VI on this forum.
> Hope this helps.
> John Wilson
> Sanders Engineering & Analytical Services, Inc.
> Mobile, Alabama
> 251-633-4120
> johnwilson@pctechnician.net
>
> ------------------------------------------------------------------------
> Name: peak_detection.vi
> peak_detection.vi Type: LabVIEW Instrument (application/x-unknown-conten
t-type-LabVIEWInstrument)
> Encoding: base64
> Description: Peak Detection Program