I was wondering if there was a simpler/faster way of calculating the area between the two curves at each point (both of which are normailsed so the total area is 1). I am currently using a simple for loop to get the area between each set of points for the two curves, then subtracting it to get the difference i.e.
for i = 1:length(normpeak)-1
resN(i) = trapz(time(i:i+1),peak1(i:i+1));
resF(i) = trapz(time(i:i+1),peak2(i:i+1));
end
res = resN-resF;
This works, but unfortunately it is very slow. I was reading up on the trapz function and from what I understood it calculates the total integral by performing something very similar to what i'm doing now (see image) but sums the data at the end.
Is there a way to stop trapz from performing this summing of the data, so I can speed up my code?

 Réponse acceptée

Star Strider
Star Strider le 23 Avr 2015

0 votes

Use the cumtrapz function to get the cumulative integral at every point.
I don’t have your data so I can’t test it, but this approach could work:
res = cumtrapz(time, peak1) - cumtrapz(time, peak2);

Plus de réponses (1)

Binbin Qi
Binbin Qi le 23 Avr 2015

0 votes

if you can get the points of every curve, you can use polyarea function to get the area, and then diff

1 commentaire

Christian Reece
Christian Reece le 23 Avr 2015
Thanks for the answer, but this doesn't seem to work. I think I was too vague before but what I need is the difference in area at each point on the curve.

Connectez-vous pour commenter.

Catégories

Tags

Community Treasure Hunt

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

Start Hunting!

Translated by