How to validate high pass filtered images with original image?
1 view (last 30 days)
Show older comments
I want to compare my original image with the high pass filtered image.Here I want to validate the high pass filtered image using a measurement.Can you suggest any specific method to validate it?
As an example how to identify illumination removal of an image after high pass filtering?
0 Comments
Answers (1)
Bjorn Gustavsson
on 23 Jan 2019
Well "validate" a high-pass filtered image to the original image is a slightly confusing concept to me - by high-pass filtering you make a signifficant change to your image. Why not start with something simple:
subplot(2,2,1)
imagesc(Orig_IM)
subplot(2,2,2)
imagesc(hpfiltered_IM)
subplot(2,2,3)
imagesc(Orig_IM-hpfiltered_IM)
subplot(2,2,2)
imagesc(log(abs(fftshift(fft2(Orig_IM-hpfiltered_IM)))))
Then you can "validate" visually - perhaps good enough, ought to be for image processing objectives, you do this type of filtering "for the eye"...
HTH
7 Comments
Bjorn Gustavsson
on 23 Jan 2019
Well, then go ask your supervisor what he means with validate - you clearly don't know, we others guess around. I've coughed up a couple of suggestions. The thing here is that, and this is explicit and concious repetition, homomorphic high-pass filtering is something you do "for the eye", and it is best to the "validation" (whatever that means) by displaying the images, their differences, and a couple of different transformed images (fft-spectrum, histogram equalized...) and see what happens.
HTH - hope this helps
See Also
Products
Community Treasure Hunt
Find the treasures in MATLAB Central and discover how the community can help you!
Start Hunting!