imagesc after fft2

Hello - im new in here :)

I've got some kind of radar's signal that I want to show with the imagesc() function.

After an fft2 I wanted to show my signal with:

figure; imagesc(abs(tabsygn-mean(y,2)));

It all worked, however when I went to reduce my axis size, something strange happened. It changed the values, however the image itself didn't change. I've used the command below:

imagesc(0:200,250:400,(abs(tabsygn-mean(y,2))));

The image on the left is original, and on the right is after the command above.

[Image Analyst]

So you're wondering about how the black spec at row 320, column 740 is missing from the image on the left? Is that it? When you say the image itself, do you mean the displayed image or the underlying image variable?

[Michal Karwacki]

No, black dot appearded when I exported the figure into jpg. I wonder how to cut out the yellow line and create image with y(150:450) and x(0:900) from the original image

[Adam]

The arguments you pass as x and y to imagesc to not change the image at all, they just give the values for the x and y data to put on the axes.

If you want to clip an image you do this in the normal way, after plotting it, using

doc xlim
doc ylim

or

doc axis

Answer 1

[Image Analyst]

Try

yourImage = abs(tabsygn-mean(y,2));
% Extract only rows 150-450 and columns 1-900.
yourImage = yourImage(150:450, 1:900);

[Michal Karwacki]

It's 2-D Matrix - there is an error with the code above.

if true
>> yourImage = abs(tabsygn-mean(y,2 ));
>> yourImage = yourImage(1:900, 150:450 ); 
Index exceeds matrix dimensions.
end

[Adam]

Well, you swapped around the indexing for some reason.

[Michal Karwacki]

well i feel stupid now. It all worked - thanks a lot

[Image Analyst]

Well, it's a common beginner mistake so you just joined thousands of other who do the same thing all the time.

Beginners think matrices are indexed m(x,y), but they are NOT.

They are indexed m(y, x) because rows are y and columns are x, and it's the rows that come first in the index list.