gray image to 8 bit planes using bit plane slicing

i have written a code to find the 8 bit planes of the gray image. i saved those images which should be in binary now. but when i am reading those images it is showing its pixel values few 0 and most 255 in binary only 0's and 1's should be there and when i did this size(d) it displayed 598 931 3.

i want it to be a in a form of 2-d array matrix with only 0's and 1's

can any one tell me what is the problem occurring?

A=imread('boy.tif');
B=bitget(A,1); figure, imshow(logical(B));title('Bit plane 1');
B=bitget(A,2); figure, imshow(logical(B));title('Bit plane 2');
B=bitget(A,3); figure, imshow(logical(B));title('Bit plane 3');
B=bitget(A,4); figure, imshow(logical(B));title('Bit plane 4');
B=bitget(A,5); figure, imshow(logical(B));title('Bit plane 5');
B=bitget(A,6); figure, imshow(logical(B));title('Bit plane 6');
B=bitget(A,7); figure, imshow(logical(B));title('Bit plane 7');
B=bitget(A,8); figure, imshow(logical(B));title('Bit plane 8');

this what i used then gave names to each of them

and when i read d=imread('bp0.tif') its giving 0 and 255 (only 0 and 255) i want ones and zeros and size should be a 2-d array why does it show 598 931 3

[Gholamreza Jahangiri]

You should first turn your image to a double, using double(A), and then you ccould have your bit for each plane.

A = imread('boy.tif');
ad = double(A);
B=bitget(ad,2); figure, imshow(B);

[Walter Roberson]

Converting to double is not necessary before doing bitget() . bitget() is happy to work on uint8.

[abdul suboor]

Please help me someone.

Question = read an 8 bit gray scale image using the OpenCV function imread().

Calculate all 8 bit planes of image, and display these bit planes a grid of 4*2 using matplotlib python library. (do on python)

[Walter Roberson]

abdul, none of that appears to be a question about MATLAB. Python has a large active community that you can be talking to, somewhere else.

[Image Analyst]

Abdul, I've attached a bit plane viewer program to my answer below, and here. But it uses MATLAB, not Python and OpenCV. Maybe it will persuade you to dump Python and switch to MATLAB. ?

Answer 1

[Image Analyst]

You have a color image. I don't know what bitget() returns or means in the case of a 3 plane color image. Why don't you use rgb2gray() or extract one of the color channels?

% Get the dimensions of the image.  
% numberOfColorBands should be = 1.
[rows, columns, numberOfColorBands] = size(B);
if numberOfColorBands > 1
  % It's not really gray scale like we expected - it's color.
  % Convert it to gray scale.
  grayImage = rgb2gray(B); % Take weighted average of channels.
end

or

% Extract the individual red, green, and blue color channels.
redChannel = B(:, :, 1);
greenChannel = B(:, :, 2);
blueChannel = B(:, :, 3);

Let us know what happens after that.

Also, see my attached bit plane viewer program.

[faraz.a]

i wrote this code

>> B=imread('boy.tif');
>> % Extract the individual red, green, and blue color channels.
redChannel = B(:, :, 1);
greenChannel = B(:, :, 2);
blueChannel = B(:, :, 3);
Index exceeds matrix dimensions.

and when i wrote this one it showed nothing

>> B=imread('boy.tif');
>> % Get the dimensions of the image.  
% numberOfColorBands should be = 1.
[rows, columns, numberOfColorBands] = size(B);
if numberOfColorBands > 1
  % It's not really gray scale like we expected - it's color.
  % Convert it to gray scale.
  grayImage = rgb2gray(B); % Take weighted average of channels.
end

[Image Analyst]

Sorry - I thought you meant B has 3 planes, but you meant d. But you did not show how you used imwrite to create "bp0.tif". What array did you save? B? Do a "whos" on that variable just before you call imwrite(...,'bp0.tif').

[faraz.a]

i didn't used any command imwrite. when the bit plane images opened in the image tool i just clicked on>>file>>save as and saved the image as bpo.tif

because i need to save those bit planes so that i can use it as binary images

and sir B is the original image. d is one of the bit planes of that image. and when you meant size(B) in your code you meant the original image boy.tif right?

"You have a color image. I don't know what bitget() returns or means in the case of a 3 plane color image. Why don't you use rgb2gray() or extract one of the color channels?

% Get the dimensions of the image.  
% numberOfColorBands should be = 1.
[rows, columns, numberOfColorBands] = size(B);
if numberOfColorBands > 1
  % It's not really gray scale like we expected - it's color.
  % Convert it to gray scale.
  grayImage = rgb2gray(B); % Take weighted average of channels.
end
or

% Extract the individual red, green, and blue color channels. redChannel = B(:, :, 1); greenChannel = B(:, :, 2); blueChannel = B(:, :, 3); Let us know what happens after that."

[faraz.a]

>>   d=imread('bp0.tif');
>> whos d
Name        Size                 Bytes  Class    Attributes

d         598x932x3            1672008  uint8

[Image Analyst]

No, don't do that. File/saveas saves the whole figure as an image. You want just the bit plane, so please use only imwrite() to save bp0.tif.

[faraz.a]

okay how to use that sir imwrite(figure1)? suppose i want to save the bit plane 1 which will open in figure 1 image tool how should i write it?

A=imread('boy.tif');
B=bitget(A,1); figure, imshow(logical(B));title('Bit plane 1');

[Image Analyst]

imwrite(uint8(B), 'bp1.png');

[faraz.a]

yup working

 size(d)

1024 1536

[faraz.a]

hello sir problem again

as you told i wrote the code

 A=imread('boy.tif');
B=bitget(A,1); figure, imshow(logical(B));title('Bit plane 1');imwrite(uint8(B), 'bp0.png');
B=bitget(A,2); figure, imshow(logical(B));title('Bit plane 2');imwrite(uint8(B), 'bp1.png');
B=bitget(A,3); figure, imshow(logical(B));title('Bit plane 3');imwrite(uint8(B), 'bp2.png');
B=bitget(A,4); figure, imshow(logical(B));title('Bit plane 4');imwrite(uint8(B), 'bp3.png');
B=bitget(A,5); figure, imshow(logical(B));title('Bit plane 5');imwrite(uint8(B), 'bp4.png');
B=bitget(A,6); figure, imshow(logical(B));title('Bit plane 6');imwrite(uint8(B), 'bp5.png');
B=bitget(A,7); figure, imshow(logical(B));title('Bit plane 7');imwrite(uint8(B), 'bp6.png');
B=bitget(A,8); figure, imshow(logical(B));title('Bit plane 8');imwrite(uint8(B), 'bp7.png');

then i wrote this

d=imread('bp7.png'); imshow(d)

it displayed a complete blank image why is that now?

[faraz.a]

if i am using imwrite(uint8(B), 'bp0.png'); it is just saving a black image.. complete black image

[Image Analyst]

It's not, it's just that with a gray level of 0 or 1 you can't see it because it's too dark. If you looked in the variable editor, you'd see the values. Try displaying it with []:

imshow(d, []);

[faraz.a]

oh okay i got it thank you. sir i want to take pixel value from the binary image and assign it to some variable X. like this

   d=imread('bp0.png');
 i=0;j=0;  
 d(i,j)=X  .. i want to give X as the pixel value of 0th row 0th column of the binary image

i know it is not correct, then what is the correct way to do this

[faraz.a]

sir i trying to execute this but it is not displaying any value.

>> d=imread('bp5.png');
[n,m]=size(d); %n rows m is columns
i=0;j=0;cont0=0;cont1=0;cont=0;
for i=0:n; 
for j=0:m;
X=d(i,j); % X is for 1st step pixel value of 0th row and 0th column
if X==0;
cont0= 1 + cont0;
end
if X==1;
cont1= 1 + cont1;
end
cont= 1 + cont;
end_for
end_for
p0= mod(cont0,cont);
p1= mod(cont1,cont);
h= -p0*log2(p0) -p1*log2(p1);
disp(h);
disp(p0);
disp(p1);
end

[Image Analyst]

None of that is necessary. It looks like you're trying to calculate the entropy. If so, just do this:

entropyOfThisImage = entropy(d);

[faraz.a]

yes sir but in some cases i need to consider the pixel value neighbors

my program method is correct i am sure but there is some problem in if else and else if statements. i am unable to rectify it can you please tell what wrong i am doing with the if else statements please may be i am not ending the if else stements properly but unable to understand how to do please check sri

d=imread('bp5.png');
[n,m]=size(d);
i00=1;
i01=2;
i10=3;
i11=4;
c0(i00)=0;c1(i00)=0;c0(i01)=0;c1(i01)=0;c0(i10)=0;c1(i10)=0;c0(i11)=0;c1(i11)=0;count(i00)=0;count(i01)=0;count(i10)=0;count(i11)=0;cont=0;   %define
i=2; j=2; % i=rows, j= columns 
for i=2:n;
for j=2:m;
X=d(i,j);
A=d(i,j-1); 
B=d(i-1,j);
if A==0 && B==0
if X==0 
c0(i00)=1+c0(i00);
else 
c1(i00)=1+c1(i00); 
end
count(i00)=1+count(i00);
end
if A==0 && B==1
if X==0 
c0(i01)=1+c0(i01);
else 
c1(i01)=1+c1(i01); 
end
count(i01)=1+count(i01);
end
if A==1 && B==0 
if X==0  
c0(i10)=1+c0(i10);
else 
c1(i10)=1+c1(i10); 
end
count(i10)=1+count(i10);
end
if A==1 && B==1
if X==0 
c0(i11)=1+c0(i11);
else if X==1
c1(i11)=1+c1(i11);
end
count(i11)=1+count(i11);
end
cont=1+cont;
end
end
end
end 
end
% c0me steady state probabilities 
p=count/cont;
%now display these all the below conditional probabilities p0(i00)=mod(c0(i00),cont(i00))
p0=c0/count;
p1=c1/count;
h= -p0*log2(p0)- p1*log2(p1);
E=sum(p*h)
end
end
end
end

Answer 2

[humbertinnho]

Ps.: Img is a 2D Image (gray colors only):

function y = Linear_Bit( Img )

b1 = double(bitget(Img,1));
b2 = double(bitget(Img,2));
b3 = double(bitget(Img,3));
b4 = double(bitget(Img,4));
b5 = double(bitget(Img,5));
b6 = double(bitget(Img,6));
b7 = double(bitget(Img,7));
b8 = double(bitget(Img,8));

Img_b = cat(8,b1,b2,b3,b4,b5,b6,b7,b8);

figure,
imshow(Img), title('Original:');

figure, 
subplot(2,2,1)
imshow(b1), title('Bit Plan: 1');
subplot(2,2,2)
imshow(b2), title('Bit Plan: 2');
subplot(2,2,3)
imshow(b3), title('Bit Plan: 3');
subplot(2,2,4)
imshow(b4), title('Bit Plan: 4');
figure, 
subplot(2,2,1)
imshow(b5), title('Bit Plan: 5');
subplot(2,2,2)
imshow(b6), title('Bit Plan: 6');
subplot(2,2,3)
imshow(b7), title('Bit Plan: 7');
subplot(2,2,4)
imshow(b8), title('Bit Plan: 8');

y = Img_b;

If you want to add Planes, u can use this:

Img_b = uint8(Linear_Bit(Img));
b1 = Img_b(:,:,1);
b2 = Img_b(:,:,2)*2;
b3 = Img_b(:,:,3)*4;
b4 = Img_b(:,:,4)*8;
b5 = Img_b(:,:,5)*16;
b6 = Img_b(:,:,6)*32;
b7 = Img_b(:,:,7)*64;
b8 = Img_b(:,:,8)*128;

...and just add like

New_Image = b7 + b8;
imshow(New_Image);

Answer 3

[Sufyan Parkar]

A=imread('boy.tif');
B=rgb2gray(A); %..............as for me i took a color image so i had to do rgb2gray
% the workspace shows the dimensions of the image and mine was 435x580 and the class of my grayscale image was uint8 thus it required 8bits
%hence to extract the plane i performed bitand of B and the binary equivalent of the required plane
for i=1:435
    for j=1:580
        MSB(i,j)=bitand(B(i,j),bin2dec('10000000'));
        LSB(i,j)=bitand(B(i,j),bin2dec('00000001'));
        2nd(i,j)=bitand(B(i,j),bin2dec('01000000'));
        3rd(i,j)=bitand(B(i,j),bin2dec('00100000'));
        4th(i,j)=bitand(B(i,j),bin2dec('00010000'));
        5th(i,j)=bitand(B(i,j),bin2dec('00001000'));
        6th(i,j)=bitand(B(i,j),bin2dec('00000100'));
        7th(i,j)=bitand(B(i,j),bin2dec('00000010'));
    end
end
figure,imshow(MSB);
figure,imshow(LSB);
figure,imshow(2nd);
figure,imshow(3rd);
figure,imshow(4th);
figure,imshow(5th);
figure,imshow(6th);
figure,imshow(7th);

%this will give you all the 8bit-plane

[Vivek]

No this code gives an error in MATLAB.

[Image Analyst]

@Vivek you're right. This is a poorly written program that won't even run. You can't start variable names with a number. Here it is less terrible, but really, you should be using my method above instead of this:

rgbImage = imread('peppers.png');
grayImage = rgb2gray(rgbImage); %..............as for me i took a color image so i had to do rgb2gray
subplot(3, 3, 1);
imshow(grayImage);
% The workspace shows the dimensions of the image and mine was 435x580 and the class of my grayscale image was uint8 thus it required 8bits
% hence to extract the plane i performed bitand of B and the binary equivalent of the required plane
MSB=bitand(grayImage,bin2dec('10000000'));
LSB=bitand(grayImage,bin2dec('00000001'));
plane2=bitand(grayImage,bin2dec('01000000'));
plane3=bitand(grayImage,bin2dec('00100000'));
plane4=bitand(grayImage,bin2dec('00010000'));
plane5=bitand(grayImage,bin2dec('00001000'));
plane6=bitand(grayImage,bin2dec('00000100'));
plane7=bitand(grayImage,bin2dec('00000010'));

% Display the bit planes.
subplot(3, 3, 2); imshow(LSB); title('Bit Plane 1')
subplot(3, 3, 3); imshow(plane2); title('Bit Plane 2')
subplot(3, 3, 4); imshow(plane3); title('Bit Plane 3')
subplot(3, 3, 5); imshow(plane4); title('Bit Plane 4')
subplot(3, 3, 6); imshow(plane5); title('Bit Plane 5')
subplot(3, 3, 7); imshow(plane6); title('Bit Plane 6')
subplot(3, 3, 8); imshow(plane7); title('Bit Plane 7')
subplot(3, 3, 9); imshow(MSB); title('Bit Plane 8')

% This will give you all 8 bit-planes.

Here is my method, as well as being attached.

% Demo to extract and view bitplanes in a gray scale image.
clc;    % Clear the command window.
workspace;  % Make sure the workspace panel is showing.
fontSize = 20;
% Read in a standard MATLAB gray scale demo image.
folder = fullfile(matlabroot, '\toolbox\images\imdemos');
button = menu('Use which demo image?', 'CameraMan', 'Moon', 'Eight', 'Coins', 'Pout');
if button == 1
    baseFileName = 'cameraman.tif';
elseif button == 2
    baseFileName = 'moon.tif';
elseif button == 3
    baseFileName = 'eight.tif';
elseif button == 4
    baseFileName = 'coins.png';
else
    baseFileName = 'pout.tif';
end
% Read in a standard MATLAB gray scale demo image.
folder = fullfile(matlabroot, '\toolbox\images\imdemos');
% Get the full filename, with path prepended.
fullFileName = fullfile(folder, baseFileName);
if ~exist(fullFileName, 'file')
    % Didn't find it there.  Check the search path for it.
    fullFileName = baseFileName; % No path this time.
    if ~exist(fullFileName, 'file')
        % Still didn't find it.  Alert user.
        errorMessage = sprintf('Error: %s does not exist.', fullFileName);
        uiwait(warndlg(errorMessage));
        return;
    end
end
grayImage = imread(fullFileName);
% Get the dimensions of the image.  numberOfColorBands should be = 1.
[rows columns numberOfColorBands] = size(grayImage);
% Display the original gray scale image.
subplot(2, 2, 1);
imshow(grayImage, []);
title('Original Grayscale Image', 'FontSize', fontSize);
% Enlarge figure to full screen.
set(gcf, 'Position', get(0,'Screensize')); 
% Let's compute and display the histogram.
[pixelCount grayLevels] = imhist(grayImage);
subplot(2, 2, 2); 
bar(pixelCount);
grid on;
title('Histogram of Original Grayscale Image', 'FontSize', fontSize);
xlim([0 grayLevels(end)]); % Scale x axis manually.
% Let's use this spot for the output image.
subplot(2, 3, 5); 
for bp = 0 : 7
    % Compute the bit plane image.
    bitPlaneImage = bitget(grayImage, bp+1);
    %     bitPlaneImage = bitand(grayImage, 2^bp);
    % Display it as pure black and white (not gray scale).
    imshow(bitPlaneImage, []);
    caption =  sprintf('You are now viewing bitplane %d.', bp);
    title(caption, 'FontSize', fontSize);
    % If it's not the last bit plane, ask them if they want to continue.
    if bp ~= 7
        message = sprintf('%s\nDo you want to continue', caption);
        button = questdlg(message, 'Continue?', 'Yes', 'No', 'Yes');
        drawnow;	% Refresh screen to get rid of dialog box remnants.
        if strcmpi(button, 'No')
            break;
        end
    end
end
msgbox('Done with demo');

[Vivek]

Thank you. I used your code but how can I save the all bitplane extracted images? So I convert this code into simple form which is look like this.

[Image Analyst]

You can use imwrite if you want to save them to a disk file.

Otherwise (less likely) if you want to save all the individual arrays to a variable for some reason (instead of just using bitPlaneImage immediately in the loop), then you can write them to a cell array

https://matlab.fandom.com/wiki/FAQ#What_is_a_cell_array?

% Compute the bit plane image and save as a cell in an 8 element cell array.
bitPlaneImage{bp} = bitget(grayImage, bp+1);
%     bitPlaneImage = bitand(grayImage, 2^bp);
% Display it as pure black and white (not gray scale).
imshow(bitPlaneImage{bp}, []);