bit plane not stored correctly

hello i'm a bit stuck with this program, the goal is to store different slices of 3 different images into a matrix, this works except for slice 16, since the bitvalues were low i have upgraded them otherwise the program sees them as zeros and the image is black. The bits are changed correcly when i load the image seperate but when i use imshow to display the different slices the 16th ( also 8th) slice is black.

What do i do wrong , i also tried ai but to no avail i hope someone can help me

ps im still a greenhorn at this just started..

this is the program

% Load existing 256x256 8bit grayscale images
image1 = imread('baboon.bmp'); % Ensure '8bit_1.bmp' is in the current directory
image2 = imread('Grayscale-Lena-Image.png'); % Ensure '8bit_2.bmp' is in the current directory
image3 = imread('house 256_ 8bit.ppm');

imshow('house 256_ 8bit.ppm');

% Display histogram of the image
figure;
itemp=image3;
imhist(itemp);
title('Histogram of the Image');


% Check if images are grayscale
if size(image1, 3) ~= 1 || size(image2, 3) ~= 1
    error('Images must be grayscale.');
end

% Store the image matrices in a cell array
images = {image1, image2, image3};

% Initialize a 3D matrix to hold the bit planes (16 for 8 bits per image)
bit_planes_3d = zeros(size(image1, 1), size(image1, 2), 16);

% Extract bit planes from the first image (1-8)
for i = 1:8
    bit_planes_3d(:, :, i) = bitget(images{1}, i);
end

% Extract bit planes from the second image (9-16)
for i = 1:7
    bit_planes_3d(:, :, i + 8) = bitget(images{2}, i);
end
%  Extract the 8th bit plane and modify the pixel values for i = 1:r for j = 1:c
%  Example modification: set the 8th bit to 1 if the current pixel value is above a 
%  threshold if itemp(i, j) 
%  > 128 itemp(i, j) = bitset(itemp(i, j), 8, 1); else itemp(i, j) = bitset(itemp(i, j), 8, 0); 
%  end end end

for i = 8
    it= image3;   %read the image
    itemp2 = it(:,:,1);
    [r,c]= size(itemp2);         % get the dimensions of image
    s = zeros(r,c,8);
    % pre allocate a variable to store 8 bit planes of the image      % nb X = zeros(2,3,4)= 2x3x4 matrix/ array
    if itemp2(r, c) > 128 itemp2(r, c) = bitset(itemp2(r, c), 8, 1);  %bit 8 word op 1 gezet
    else itemp2(r, c) = bitset(itemp2(r, c), 8, 0); 
        bit_planes_3d(:, :, i + 8) = bitget(itemp2, i);
    end
end

%Display the bit planes
%Open a new figure window
for i = 1:16
    subplot(4, 4, i); % 4 rows, 4 columns for 16 bit planes
    imshow(bit_planes_3d(:, :, i));
    title(['Bit Plane ' num2str(i)]);
end

[DGM]

I'm not sure where this is going, but this is why slice 16 starts out blank

% Extract bit planes from the second image (9-16)
for i = 1:8 % <-- any reason why you're skipping the MSB?
    bit_planes_3d(:, :, i + 8) = bitget(images{2}, i);
end

Answer 1

[Sameer]

Hi @Kees

The issue you're facing with the 8th and 16th bit planes being black might be due to a few reasons:

• The loop for extracting bit planes from the second image should go from 1 to 8, not 1 to 7. This might be why the 16th slice is not being processed correctly.
• Ensure that the bit plane extraction logic is correctly implemented for all images. The current logic for modifying the 8th bit plane seems to be inside a loop that only runs for i = 8, which might not be correct.

Here’s a revised version of your code with these corrections:

image1 = imread('baboon.bmp'); 
image2 = imread('Grayscale-Lena-Image.png');
image3 = imread('house 256_ 8bit.ppm');
imshow('house 256_ 8bit.ppm');

% Display histogram of the image
figure;
itemp = image3;
imhist(itemp);
title('Histogram of the Image');

% Check if images are grayscale
if size(image1, 3) ~= 1 || size(image2, 3) ~= 1
    error('Images must be grayscale.');
end

% Store the image matrices in a cell array
images = {image1, image2, image3};

% Initialize a 3D matrix to hold the bit planes (16 for 8 bits per image)
bit_planes_3d = zeros(size(image1, 1), size(image1, 2), 16);

% Extract bit planes from the first image (1-8)
for i = 1:8
    bit_planes_3d(:, :, i) = bitget(images{1}, i);
end

% Extract bit planes from the second image (9-16)
for i = 1:8
    bit_planes_3d(:, :, i + 8) = bitget(images{2}, i);
end

% Modify the 8th bit plane of the third image
itemp2 = image3(:,:,1);
[r, c] = size(itemp2);
for i = 1:r
    for j = 1:c
        if itemp2(i, j) > 128
            itemp2(i, j) = bitset(itemp2(i, j), 8, 1);
        else
            itemp2(i, j) = bitset(itemp2(i, j), 8, 0);
        end
    end
end

% Extract the modified 8th bit plane from the third image
bit_planes_3d(:, :, 16) = bitget(itemp2, 8);

% Display the bit planes
figure;
for i = 1:16
    subplot(4, 4, i); % 4 rows, 4 columns for 16 bit planes
    imshow(bit_planes_3d(:, :, i));
    title(['Bit Plane ' num2str(i)]);
end

Hope this helps!

[Kees]

i've implemented your points, but unfortunate the program does not work for slice 8, and 16

[Sameer]

Could you attach the images so that I can check on my end?

[Kees]

i attached the images, if i use baboon for the 3rd the problem remains

Answer 2

[Kees]

i might just found the answer the 3rd picture is ppm this is not supported

[DGM]

Imread() does support PPM.

unzip pep_u16.ppm.zip % for the forum

% this is a simple uint16 RGB PPM
inpict = imread('pep_u16.ppm');
imshow(inpict)

Are you sure there isn't some other issue with the file being read incorrectly? IIRC, there are ways a PPM might be configured such that imread() might not return values as expected without extra work.

[Kees]

hi if this is the case why does it work for the other slides?

[DGM]

I'm not sure what part isn't working. This specific objection is regarding the reading of a PPM file. I don't have the file, so I can't say why a particular PPM file might not be read correctly. All I'm saying is that it should be possible.

As to why there's a problem with image3, I'm also not sure, since I thought this problem was about the initialization of the working array from the MSB bitplane of image2. Like I said, I don't know where this is going, so I can't comment on what the processing of image3 is intended to produce.

[Kees]

ok wel thanx for your input, @ sameer i attached the pictures