Chessboard illumination - Problems using imrotate

Hello! I'm doing a chessboard with dimensions of 512x512. In my code I should apply a parameter which makes the illumination of the chessboard varies given an angle (MCangle). I used the imrotate function, but the output image shows the chessboard with black corners. I wanted that my rotate image fill my original image. I've tried to use imrotate without the 'crop' parameter, but it shows an error saying "Matrix dimensions must agree.", as expected. Can someone help me with this problem? I would be very thankful!! And sorry if my english is bad.

f1=255*ones(64);
f2=zeros(64);
R1=[f1 f2 f1 f2 f1 f2 f1 f2];
R2=[f2 f1 f2 f1 f2 f1 f2 f1];
chess=[R1;R2;R1;R2;R1;R2;R1;R2];

vl=1:512;
vc=ones(512,1);
MA=vc*vl;
MB=mat2gray(MA);

%illumination intensity
MC=MB*(0.6)+0.2;%MC=MB*(IMAX-IMIN)+IMIN;

MCangle=imrotate(MC,45,'crop');

chessintensity=chess.*MCangle;

chessI8=uint8(chessintensity);

imshow(chessI8)

[Matt J]

I took the liberty of editting your post and running your code in it. As you can see, it runs error-free.

[Joana Teixeira]

Yes, but I don't get exactly what I want.

[Matt J]

Well, you need to edit your post to describe what the actual problem is. Currently, you describe a "matrix dimensions must agree" problem that doesn't exist.

[Joana Teixeira]

"I used the imrotate function, but the output image shows the chessboard with black corners. I wanted that my rotate image fill my original image." I dont want the black corners.

Answer 1

[Matt J]

f1=255*ones(64);
f2=zeros(64);
R1=[f1 f2 f1 f2 f1 f2 f1 f2];
R2=[f2 f1 f2 f1 f2 f1 f2 f1];
chess=[R1;R2;R1;R2;R1;R2;R1;R2];

[x,y]=ndgrid((1:512)-mean(1:512));
MB=rescale(-x*sind(45)+y*cosd(45),0,1);

%illumination intensity
MCangle=MB*(0.6)+0.2;%MC=MB*(IMAX-IMIN)+IMIN;

chessintensity=chess.*MCangle;

chessI8=uint8(chessintensity);

imshow(chessI8)

[Joana Teixeira]

It worked!! Thank you so much!! Can you tell what the 0 and 1 do in the rescale function?

[Matt J]

rescale(A,0,1)

will rescale A so that min(A)=0 and max(A)=1.