how do i perform local normalization of fingerprint image. please correct this code the sise of my image is 480x640

clc;   
I1=imread('D:/aa.png');
I1=imresize(I1,[450,600])
I= im2double(I1);
set(gcf, 'units','normalized','outerposition',[0 0 1 1]);
drawnow;
[rows columns numberOfColorBands] = size(I)
disp([rows columns numberOfColorBands]);
blockSizeR = 150; % Rows in block.
blockSizeC = 100; % Columns in block.
plotIndex = 1;
numPlotsR = size(ca, 1);
numPlotsC = size(ca, 2);
for r = 1 : numPlotsR
  for c = 1 : numPlotsC
      fprintf('plotindex = %d,   c=%d, r=%d\n', plotIndex, c, r);
      % Specify the location for display of the image.
      subplot(numPlotsR, numPlotsC, plotIndex);
              rgbBlock = ca{r,c};
      imshow(rgbBlock); % Could call imshow(ca{r,c}) if you wanted to.
      [rowsB columnsB numberOfColorBandsB] = size(rgbBlock);
             caption = sprintf('Block #%d of %d\n%d rows by %d columns', ...
          plotIndex, numPlotsR*numPlotsC, rowsB, columnsB);
      title(captio);
      drawnow;
             plotIndex = plotIndex + 1;
  end
end
subplot(4, 6, 1);
imshow(I);
title('Original Image');
*m1=128;
v1=128*128;
for l=1:8
  for k=1:8
           m = mean(ca{l});  
            v(k)=var(ca{l}(:));
          coeff(k)=v(k)/v1;
  end  
end
for k=1:8
for i=1:120
      for j=1:160
          norimg(i, j) =  m1 + coeff(k) .* (I1(i, j) - m(k) );
            end
end
end*
_for i=1:120
      for j=1:160
          figure,imshow(norimg(i, j));
      end
end
_

[Walter Roberson]

Why are you trying to show 120 times 160 images? Why are you not using just a simple

imshow(norimg)

with no loop?

[Image Analyst]

And why are you referring to ca{l} when ca is a 2D array? Plus other errors too numerous to detail. I think you had better think this over a lot and then post again later.

[Walter Roberson]

"ca" is undefined

[Image Analyst]

Yes, it's a version of my "split an image up into blocks" code I've posted before but he pretty much didn't understand it, and then it just got worse when he tried to adapt it, like when he took out the line:

ca = mat2cell(rgbImage, blockVectorR, blockVectorC, numberOfColorBands);

I'd help but I'm not sure what he wants to do. I'd have to study it more to figure it out and fix it, but I don't have time tonight.

Answer 1

[Thorsten]

It's a one-liner

X = blkproc(I, [150 100], @(x) (0.5 + x - mean2(x))./(std2(x)/0.25));

You may want to change the normalizing function. Note that after im2double the image values are mapped form [0, 255] to the range [0, 1], that's why I have replaced your 128 values by 0.5.