How to perform 2D curve fitting on an image

Hello all,

I am very new to image processing and am struggling with post processing some data I have gathered. I have an image of some experimental data with frequency accross the x axis and wavenumber accrosss the y axis. Image attached for you to view..

It clearly can be seen that there are various curves (about 5 of them) that vary accross this 3D space, due to their large amplitude.

I would like to be able to curve fit data in terms of y = f(x) (wavenumber = f(frequency)) at the pixels of max amplitude. Does anyone have an idea of how to do this?

I thought I could loop through each column and find the peaks in order to determine the function I desire but I don't feel that this is too elegant.

Keen to hear if anyone has any other smart ideas

Regards,

Euan

[Addy]

I am working with ultrasonic guided waves. Looks like you are trying to achieve mode filterning based on wavenumber plot.

Answer 1

[Vinoth Venkatesan]

Assuming you want to process the plot as an image, this is a possible workflow:

• Binarize the grayscale image ("rgb2gray" and "imbinarize")
• Find the connected components in the binarized image ("bwconncomp")
• Use the 'PixelIdxList' that you get for these components to fit a curve (Interactively using "cftool" or programmatically using functions in the Curve Fitting Toolbox)

You might have to do some post-processing to convert the 'PixelIdxList' to [x,y] data (the "ind2sub" function might be of use here). Refer to these doc links for more info on these functions:

https://www.mathworks.com/help/matlab/ref/rgb2gray.html

https://www.mathworks.com/help/images/ref/imbinarize.html

https://www.mathworks.com/help/images/ref/bwconncomp.html

https://www.mathworks.com/help/curvefit/curve-fitting.html