Curve at the peak values

1 次查看(过去 30 天)
Mepe
Mepe 2019-3-16
评论: dpb 2019-3-17
For the evaluation of measurement results, a curve should be laid through the maximum values (polynomial or spline). Is there a direct possibility here or must the maximum measured values be read out separately?
Many thanks!
Picture1.png
  3 个评论
dpb
dpb 2019-3-17
[OP's Answer moved to Comment...dpb]
Thanks, with
[pkt,lct] = findpeaks(ChannelE,Frequency,'Threshold',0.8);
this works quite well. What's the best way to create a fitting curve through the points? With polyfit or lsqcurvefit I was not very successful...
Many thanks!
dpb
dpb 2019-3-17
A smoothing spline would likely be the ticket there -- or a N peak mixed Gaussian or somesuch depending upon whether there is some sort of expected shape based on whatever it is the data represent and what the bounding curve is for--altho there's a dichotomy in that the two major peaks with the example "fit" are treated as one while the others with similar spacing are separated. The only difference appears to be the two are of essentially the same magnitude.

请先登录,再进行评论。

采纳的回答

Mepe
Mepe 2019-3-17
Thanks, with
[pkt,lct] = findpeaks(ChannelE,Frequency,'Threshold',0.8);
this works quite well. What's the best way to create a fitting curve through the points? With polyfit or lsqcurvefit I was not very successful...
Many thanks!

更多回答(0 个)

类别

Help CenterFile Exchange 中查找有关 Interpolation 的更多信息

Community Treasure Hunt

Find the treasures in MATLAB Central and discover how the community can help you!

Start Hunting!

Translated by