Mathematics Stack Exchange is a question and answer site for people studying math at any level and professionals in related fields. Join them; it only takes a minute:

Sign up
Here's how it works:
  1. Anybody can ask a question
  2. Anybody can answer
  3. The best answers are voted up and rise to the top

I have a set of graphs with a large primary peak and a secondary peak that intersects the first and I want to fit two curves (one per peak) and find the point at which the two curves intersect for thresholding reasons.

The problem is that the way the curves are fit has to be exactly repeatable and algorithmic in order to reduce my error margin for comparison purposes. Is there a curve fitting method that returns a set of curves, one for each peak found?

If it helps, I'm writing this in IDL.

share|cite|improve this question
up vote 1 down vote accepted

You can fit your data to a functional form you choose. One possibility would be the sum of two Gaussian distributions plus a background: $n(x)=a+b \exp((x-\mu_1)^2/\sigma_1^2)+c \exp((x-\mu_2)^2/\sigma_2^2)$ You can use whatever peaked function you like in place of the Gaussian. Then feed it to a multidimensional minimizer to estimate the parameters. Chapter 10 of Numerical Recipes or any numerical analysis text will give you information. This will be repeatable, but the intersection point may well depend upon the functional form you choose.

share|cite|improve this answer
Thanks very much, my supervisor had suggested the sum of gaussians. Sorry if this is a really low-level question - mathematics isn't really my forte (I'm a computer vision man normally) – Nick Udell Nov 29 '11 at 14:25

Your Answer


By posting your answer, you agree to the privacy policy and terms of service.

Not the answer you're looking for? Browse other questions tagged or ask your own question.