How to increase the resolution of a list of images

I did 'image variance' for a list of images but the region of interest shows black line (see picture 1) instead of white as in picture 2. What seems to be the problem? does it need to increase the resolution or high pass filtering?

Answer 1

[Walter Roberson]

You have a series of images that all happen to be bright in certain areas. Because they are all bright there, the variance between them is low. When you display the variance, low variance would show up dark.

Variance images highlight areas that differ a lot between images, not areas that are similar.

[BA]

@Walter Roberson does this still ok? I mean calculating the blackish line region gives right pixels.

[BA]

@Walter RobersonNot sure that would be the case. I tried with darker frames and the outcome is similar(see attached new image)

[Walter Roberson]

I do not understand what you are asking?

I do not see any problem with the initial image if you are creating a "variance image"

Your first image is a classic ultrasound sonogram shape, but your second image looks it was created with some other method. It might be ultrasound but it does not appear to be created with the typical wand probe.

[BA]

@Walter Robersondoes spatial filtering/resolution improve the display? also, I need to check other matlab code of how to plot 'variance image' on the images themselves as I did above.

[Walter Roberson]

img = imread('cameraman.tif');
img1 = img; img1(1:150, 1:20) = img1(1:150, 1:20) + 1;
img2 = img; img2(1:50, 1:50) = double(img2(1:50, 1:50)) * rand(50,50);
img3 = img; img3(100:150,1:50) = img3(100:150,1:50) * 1.5;


combined = cat(3, img, img1, img2, img3);
varimg = var(double(combined), [], 3);

figure();
subplot(2,2,1); imshow(img); title('original');
subplot(2,2,2); imshow(img1); title('variation 1');
subplot(2,2,3); imshow(img2); title('variation 2');
subplot(2,2,4); imshow(img3); title('variation 3');

figure();
imshow(varimg); title('variance image');

variance image:

Black --> places where the images did not change much

White --> places where the image schanged a lot.

Gray --> places where the images changed a little

Your test2.jpg shows white where there was text, because the text changed between images. Something also changed near (0.8, 0.8) . There was also some small change in the bottom center.

If you were to mask out the text areas before taking the variance images, you would have more emphasis on the change in the bottom center.