Superimpose two images on top of each other.
    7 visualizaciones (últimos 30 días)
  
       Mostrar comentarios más antiguos
    
Hi, I am quite new to MATLAB and not very familiar with how imregtform, imwarp, and whether I should have used "affine" or "similarity/rigid" instead to achieve my goal.
I am trying to superimpose my movingRegistered image directly on top of my fixed image, but this is the outcome that I get :

I can't get my movingRegistered image to translate directly to the top of my fixed image.
Would switching my approach from intensity-based registration to control point registration/feature detection and matching be the better approach?
This is my code that I am currently using:
function myfunc()
    clear;
    clc;
    cla;
    close all
    fixed = image_resizing("test1.1.JPG");
    moving = image_resizing("test2.1.JPG");
    fixed = rgb2gray(fixed);    
    moving = rgb2gray(moving);
    fixed = imresize(fixed,0.6, 'AntiAliasing', true);    
    moving = imresize(moving, 0.6, 'AntiAliasing',true);
    [optimizer, metric] = imregconfig('monomodal');
    optimizer.MinimumStepLength = 5e-6;
    optimizer.MaximumIterations = 300; 
    tform = imregtform(moving, fixed, 'affine', optimizer, metric,'DisplayOptimization',false);
    movingRegistered = imwarp(moving,tform,'OutputView',imref2d(size(fixed)));
    imshowpair(fixed, movingRegistered,'Scaling','joint')
end    
These are my two images that I would be using  :  &
&
 &
&
Thank you so much for your help!
1 comentario
  Walter Roberson
      
      
 el 1 de Jul. de 2021
				In some recent work I was doing, I found that using Phase Correlation worked much better than any of the other ones I tested.
Respuestas (1)
  Bjorn Gustavsson
      
 el 1 de Jul. de 2021
        If your images are of equal size you can put them into different RGB-layers. Something like this:
merged(:,:,3) = fixed;
merged(:,:,1) = movingRegistered;
That should give you the fixed image as blue, and the movingRegistered as red. You might need to check the data-types (if double you'll need to normalize the merged to 0-1).
HTH
0 comentarios
Ver también
Categorías
				Más información sobre Matched Filter and Ambiguity Function en Help Center y File Exchange.
			
	Community Treasure Hunt
Find the treasures in MATLAB Central and discover how the community can help you!
Start Hunting!


