can u make black color region in to some other color say red or green in the given image but dont change the white color?

here i have attached the image in which i want green or red color instead of black color region for my academic work. please give me accurate answer.

https://mail.google.com/mail/?ui=2&ik=bcbccaefc4&view=att&th=131d66f90017f805&attid=0.1&disp=inline&realattid=f_grfxa06t0&zw

thanks in advance

[Friedrich]

I can't access the image. Please upload it somewhere else.

[somasekar jalari]

https://mail.google.com/mail/?ui=2&ik=bcbccaefc4&view=att&th=131d66f90017f805&attid=0.1&disp=inline&realattid=f_grfxa06t0&zw

copy and paste the above address , we ll get the image. still if you fail to get image let me inform the procedure to upload the image please.

thanks

[Friedrich]

The above link requieres a google mail account. You can use http://tinypic.com/ to upload pictures.

[somasekar jalari]

hello sir, i upload image according to ur suggestion.if u failed to get my image please let me know ur mail id so that i can send to u. bec this very imp to my project work

thanks in advance

[Jan]

If your project is important (for you), take the time to post the link to the image here...

Which format does your picture have? A DOUBLE or UINT8 RGB array? A greyscale image?

Do you want to replace only exact black pixels with the value [0,0,0] or all very dark colors?

@native speakers: Is "please give me accurate answer" polite? Is it inpolite to give an inaccurate answer?

[Image Analyst]

I didn't take it as impolite. I took it as just typical, normal non-native wording like you often see from non-native speakers.

[somasekar jalari]

i have binary image sir.i detected some portion from the original image now i want to make black color region to pure red or green color.help me sir.

thanks in advance

[Jan]

Please post more details by editing your original question: Does "binary image" mean a LOGICAL array? If you "detect some portion", in which format is this portion stored?

The more details, the more likely is a meaningful answer. Otherwise we waste time with guessing.

@Image Analyst: Thanks. Converting a message twice by non-native speakers with different mother-tongues leads to strange results, usually.

[Walter Roberson]

http://www.mathworks.com/matlabcentral/answers/7924-where-can-i-upload-images-and-files-for-use-on-matlab-answers

Answer 1

[Image Analyst]

OK, you say in a comment that you already have the binary image and you want the black parts of it to be red (or green). Something like this should do it:

% Extract the individual red, green, and blue color channels.
redChannel = rgbImage(:, :, 1);
greenChannel = rgbImage(:, :, 2);
blueChannel = rgbImage(:, :, 3);
% Make the black parts pure red.
redChannel(~binaryImage) = 255;
greenChannel (~binaryImage) = 0;
blueChannel (~binaryImage) = 0;
% Make the above lines 0, 255, and 0 if you want green instead.
% Now recombine to form the output image.
rgbImageOut = cat(3, redChannel, greenChannel, blueChannel);

[somasekar jalari]

thanks a lot sir. i executed your code sir but its working for rgb images+binary.but i want only for binary image.

i have binary image and in that i want to make black region into red or green color sir. please advice me.

Answer 2

[Jan]

If you have an UINT8 RGB image and want to replace [0,0,0] pixels:

rgb = uint8(floor(rand(640, 480, 3) * 10));
isBlack = all(rgb == uint8(0), 3);
r = rgb(:, :, 1);
r(isBlack) = 1;
g = rgb(:, :, 2);
g(isBlack) = 0;
b = rgb(:, :, 3);
b(isBlack) = 0;
rgb2 = cat(3, r, g, b);

Answer 3

[Daniel Shub]

While there are faster and better ways ...

x = rand(100, 100, 3);
x(1:10, 1:10, :) = 0;
y = logical([100, 100]);
z = x;

for i = 1:100
  for j = 1:100
    y(i, j) = all(x(i, j, :) == [0, 0, 0]);
  end
end

figure
image(x)

figure
image(z)

[Jan]

"all(x(i, j) == [0,0,0])" is not correct. This checks only the red channel of the 3D image array. I assume you want: "all(x(i, j, :) == [0,0,0])". But then a comparison with a scalar 0 is enough and faster. Vectorization is efficient here: replace the loops by: "all(x==0, 3)".

Answer 4

[Image Analyst]

See my color segmentation demos: http://www.mathworks.com/matlabcentral/fileexchange/?term=authorid%3A31862. With the proper method of segmentation you should be able to achieve the color replacement you want, regardless if the color is a single [r,g,b] triplet or a more spread out gamut.

Answer 5

[Andrei Bobrov]

rgb = uint8(rand(5,5,3)<.25)
rgb(rgb~=0) = uint8(randi([1 255],nnz(rgb),1))
bk = im2uint8(all(rgb==0,3))
rgb2 = rgb;

choose one from variants

% red
rgb2(:,:,1) = bk + rgb2(:,:,1)
% green
rgb2(:,:,2) = bk + rgb2(:,:,2)
% blue
rgb2(:,:,3) = bk + rgb2(:,:,3)

OR

rgb = +(rand(5,5,3)<.25)
rgb(rgb~=0) = rand(nnz(rgb),1)
bk = all(rgb==0,3)
rgb2 = rgb
rgb2(:,:,1) = rgb2(:,:,1)+bk 
image(rgb2) % red

ADD MORE (about binary image)

d = +(rand(10)<.3) %binary image
d1  = d(:,:,[ 1 1 1])
d1(:,:,1)=ones(size(d)) % red
image(d1)

[somasekar jalari]

thanks a lot sir. ur code working for binary image and got the exact result what i expected.

once again thanks.