# How can I implement block search in this program?

1 visualización (últimos 30 días)
Emmanuel el 25 de Sept. de 2014
Editada: Stephen23 el 25 de Sept. de 2014
Hello all! I wanted to compare image al , with another image b, which is a block of image form the previous image al. I tried splitting image al block by block and storing them in 4 different matrices(and implement SSD), which I've programmed below. But when I run them, it looks like its into an infinite loop. Is there any other method to process blocks and match them? Please help.
clear all;
close all;
br = uint8(zeros(125, 125, 3));
br(1:126, 1:126,:) = al(80:205, 100:225 , :)
imtool(al); imtool(br);
al= rgb2gray(al);
al = im2double(al);
br= rgb2gray(br);
br = im2double(br);
for(i =1:1:126)
for (j = 1:1:126)
c(i,j) = al(i,j)
end
end
for(i =99:1:225)
for (j = 99:1:225)
d(i,j) = al(i,j)
end
end
for(i =99:1:225)
for (j = 1:1:126)
e(i,j) = al(i,j)
end
end
for(i =1:1:126)
for (j = 99:1:225)
f(i,j) = al(i,j)
end
end
imtool(c); imtool(d); imtool(e); imtool(f)
##### 1 comentarioMostrar -1 comentarios más antiguosOcultar -1 comentarios más antiguos
Stephen23 el 25 de Sept. de 2014
You mention that you wish to "match" the blocks from the image. What exactly do you mean by this? Is pixel-wise RGB value equivalence a "match"?

Iniciar sesión para comentar.

Stephen23 el 25 de Sept. de 2014
Editada: Stephen23 el 25 de Sept. de 2014
MATLAB's feature called vectorization would be a great help, it allows you to perform operations on whole arrays, without requiring any loops. It is faster, neater, and less prone to bugs.
This means that
for(i =1:1:126)
for (j = 1:1:126)
c(i,j) = al(i,j)
end
end
can simply be
c(1:126,1:126) = al(1:126,1:126);
or, if this is the first time that c is defined:
c = al(1:126,1:126);
This also applies to the two lines
br = uint8(zeros(125, 125, 3));
br(1:126, 1:126,:) = al(80:205, 100:225 , :)
where the first line really is not required:
br = al(80:205,100:225,:);
This introduction to indexing might be useful to read too:
However you are working with image data, which often consists of 3D arrays. You will also need to consider what happens to that third array dimension.
Note that it is best to avoid i and j as a variable names, as these represent imaginary/complex values.
##### 1 comentarioMostrar -1 comentarios más antiguosOcultar -1 comentarios más antiguos
Emmanuel el 25 de Sept. de 2014
Thanks!! Now my code snippet looks like this:
c(1:126 , 1:126, :) = al(1:126 , 1:126, :)
d(1:126, 1:126,:) = al(100:225, 100:225,:)
e(1:126, 1:126,:) = al(100:225, 1:126,:)
f(1:126, 1:126,:) = al(1:126,100:225, :)
imtool(c); imtool(d); imtool(e); imtool(f)
I got four image blocks :). Is there any other efficient algorithm that will help me in block matching. Also, from where can I learn about the algorithms? Any online tutorials or sources which you might help me. I'm a beginner in image processing and matlab and would like to learn more about them

Iniciar sesión para comentar.

### Categorías

Más información sobre Neighborhood and Block Processing 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!

Translated by