How to create a checkerboard image without using the inbuilt function?
31 visualizaciones (últimos 30 días)
Mostrar comentarios más antiguos
Deep P
el 6 de Jun. de 2016
Hello All,
I wanted to create a checkerboard image(1280x960) where each color(balck or white) is 16x16 in dimension. I do not want to use inbuilt function. Please help me on this.
Thanks.
2 comentarios
Adam
el 6 de Jun. de 2016
Almost everything is an inbuilt function, but assuming you have read the help what aspect of this are you struggling with?
Using a for loop and basic matrix indexing this seems to be relatively simple to achieve.
Respuesta aceptada
Stephen23
el 6 de Jun. de 2016
Editada: Stephen23
el 6 de Jun. de 2016
Here is one way without using any toolbox functions, or any slow (and ugly) loops:
>> imgSiz = [16,12];
>> blkSiz = [4,4];
>> numRep = imgSiz./blkSiz;
>> basMat = toeplitz(mod(0:numRep(1)-1,2),mod(0:numRep(2)-1,2));
>> outImg = repelem(basMat,[blkSiz,3])
outImg =
0 0 0 0 1 1 1 1 0 0 0 0
0 0 0 0 1 1 1 1 0 0 0 0
0 0 0 0 1 1 1 1 0 0 0 0
0 0 0 0 1 1 1 1 0 0 0 0
1 1 1 1 0 0 0 0 1 1 1 1
1 1 1 1 0 0 0 0 1 1 1 1
1 1 1 1 0 0 0 0 1 1 1 1
1 1 1 1 0 0 0 0 1 1 1 1
0 0 0 0 1 1 1 1 0 0 0 0
0 0 0 0 1 1 1 1 0 0 0 0
0 0 0 0 1 1 1 1 0 0 0 0
0 0 0 0 1 1 1 1 0 0 0 0
1 1 1 1 0 0 0 0 1 1 1 1
1 1 1 1 0 0 0 0 1 1 1 1
1 1 1 1 0 0 0 0 1 1 1 1
1 1 1 1 0 0 0 0 1 1 1 1
This makes a simple binary image. Obviously you need to replicate along the third dimension if you need an RGB image.
If you don't have repelem, then this FEX submission also does the job: http://www.mathworks.com/matlabcentral/fileexchange/24536-expand
2 comentarios
Image Analyst
el 6 de Jun. de 2016
Yes, and so is repmat() which you could use if you created one white & dark pair. Maybe it's only the "checkerboard()" function that is prohibited. http://www.mathworks.com/matlabcentral/answers/38787-what-can-be-programmed-without-any-built-in-functions
Más respuestas (3)
DGM
el 20 de Feb. de 2023
Editada: DGM
el 20 de Feb. de 2023
How about using only mod(), xor(), and basic operators (colon, transpose, relational)?
% parameters
squaresize = [20 20]; % the size of squares [y x]
sizeout = [200 200]; % the image size [y x]
% create a checkerboard mask
xx = mod(0:(sizeout(2)-1),squaresize(2)*2) < squaresize(2);
yy = mod(0:(sizeout(1)-1),squaresize(1)*2) < squaresize(1);
mask = xor(xx,yy');
% display it
imshow(mask)
Okay, I guess I did use imshow(), but that's not necessary to "create" the image.
I guess I'm also relying on implicit array expansion, so if I were to answer this in early 2016, I'd have to use bsxfun() or repmat().
% generalized expansion was introduced in R2016b
mask = bsxfun(@xor,xx,yy'); % works back to R2007a
Maybe that reduces the value a bit.
While this example is more complicated than simple blockwise replication using a single 2x2 checkerboard block and repmat(), it does allow the image size to be independent of the square size without adding extra steps. This makes it simple to generalize further without the introduction of new concepts.
% parameters
squaresize = [4 4]; % the size of squares [y x]
sizeout = [20 25]; % the image size [y x]
offset = [2 2]; % offset from the NW corner [y x]
% create a checkerboard mask
xx = offset(2):(sizeout(2) + offset(2) - 1);
yy = offset(1):(sizeout(1) + offset(1) - 1);
xx = mod(xx,squaresize(2)*2) < squaresize(2);
yy = mod(yy,squaresize(1)*2) < squaresize(1);
mask = xor(xx,yy');
% display it
imshow(mask)
See also:
0 comentarios
John BG
el 7 de Jun. de 2016
Editada: John BG
el 25 de Jul. de 2016
Deep P
this works, no inbuilt functions:
sx=1280;sy=960;
bas=16 % length base square side
% assume sx and sy are multiples of base
Lx=(-1).^[1:1:sx/bas];
Ly=(-1).^[1:1:sy/bas];
A=ones(sx,sy);
[Ai Aj]=ind2sub([sx sy],[1:1:sx*sy]);
Ai2=reshape(Ai,[sx sy]);
Aj2=reshape(Aj,[sx sy]);
linex=[1:bas:sx];linex=[linex sx];
liney=[1:bas:sy];liney=[liney sy];
setx=zeros(1,bas);
for k=2:1:(length(linex)-1)
L1=[linex(k-1):1:(linex(k)-1)]
setx=[setx;L1];
end
setx(1,:)=[];
sety=zeros(1,bas);
for k=2:1:length(liney)-1
L2=[liney(k-1):1:liney(k)-1];
sety=[sety;L2];
end
sety(1,:)=[];
blk=ones(bas,bas);
for k=1:1:sx/bas-1
for s=1:1:sy/bas-1
Lx1=blk*Lx(k);
Ly1=blk*Ly(s);
L12=Lx1.*Ly1;
A(setx(k,:),sety(s,:))=A(setx(k,:),sety(s,:)).*L12;
end;
end;
check the result with the marker on
imshow(A)
If you find this answer of any help solving your question,
please click on the thumbs-up vote link, or mark this answer as accepted
thanks in advance
John
1 comentario
TEGENE Garedew
el 23 de Nov. de 2019
thanks that is nice but it doesn't display 8 squares only 7 squares and a white region on the left and bottom region
Nipuna
el 7 de Sept. de 2020
Following function worked
call the function binerycheckerboard(1280,960)
function a = binerycheckerboard(n,y)
b = ones(n,y);
for i=0:n-1
for a=0:y-1
c = rem(i,16);
d = fix(i/16);
e = rem(d,2);
f = rem(a,16);
g = fix(a/16);
h = rem(g,2);
if e == 0 && h == 0 || e ~= 0 && h ~= 0
b(i+1,a+1) = 1;
else
b(i+1,a+1) = 0;
end
end
end
imshow(b)
imtool(b)
0 comentarios
Ver también
Categorías
Más información sobre Image Processing Toolbox en Help Center y File Exchange.
Productos
Community Treasure Hunt
Find the treasures in MATLAB Central and discover how the community can help you!
Start Hunting!