how to enhance blue color alone of an image?

how to enhance blue color alone of an image?

[Jan]

What exactly does "enhance" mean?

[Dorothy Lopez]

I think he want to tone pictures to blue color images, adding more blue hues on the photos. From my knowledge, many tools can do it, famous one are Gimp, On1, PS, LR.....

Answer 1

[David Lieberman]

This question is vague, but I'll try to help. First off, don't just modify the blue sepearation. That will shift everything toward blue, including gray pixels, and I don't think that's your intent, so ignore Jan's suggestion above.

Color images are often rgb, that is img(:,:,1) is the red, img(:,:,2) is green, and img(:,:,3) is blue. By 'blue color alone', I imagine you are wanting to adjust pixels that are predominantly blue to become 'more so'.

Note: crudly, r+g+b = the gray component, cyan = g+b, magenta=r+b, and yellow=red+green

% get components r=img(:,:,1); g=img(:,:,2); b= img(:,:,3);

% When there is more blue than anything else, find out by how much extra_blue = max(b-max(r,g),0);

Now, you can try to add additional blueish component, as long as you stay in the valid range.

b_mod = min(b + k*extra_blue),1); % k is a scale of the amount of extra blue to add, say 0.1 for example

Alternatively, you can add the extra blue and scale down the non-blue components. Reduce r, g, and (b-extra_blue) by the exact same scale factor (say 0.9), that will maintain gray and all hue angles, then add a multiple of the excess blue (say 1.1) to the b channel.

There are much better ways, but then I'd have to ask you about hue angle, saturation, and it gets complicated quickly... I've said enough. I hope this helps.

Answer 2

[Image Analyst]

I think you'll find this demo illustrates it very well:

clc;    % Clear the command window.
close all;  % Close all figures (except those of imtool.)
clear;  % Erase all existing variables.
workspace;  % Make sure the workspace panel is showing.
format longg;
format compact;
fontSize = 20;

% Read in a standard MATLAB color demo image.
folder = fullfile(matlabroot, '\toolbox\images\imdemos');
baseFileName = 'football.jpg';
% 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
rgbImage = imread(fullFileName);
% Get the dimensions of the image.  numberOfColorBands should be = 3.
[rows columns numberOfColorBands] = size(rgbImage);
% Display the original color image.
subplot(2, 2, 1);
imshow(rgbImage, []);
title('Original Color Image', 'FontSize', fontSize);
% Enlarge figure to full screen.
set(gcf, 'units','normalized','outerposition',[0 0 1 1]);

% Convert to hsv color space.
hsv = rgb2hsv(rgbImage);
% Display the color channels.
hueImage = hsv(:, :, 1);
saturationImage = hsv(:, :, 2);
valueImage = hsv(:, :, 3);
subplot(2, 2, 2);
imshow(hueImage, []);
title('Hue Channel', 'FontSize', fontSize);
subplot(2, 2, 3);
imshow(saturationImage, []); 
title('Saturation Channel', 'FontSize', fontSize);
subplot(2, 2, 4);
imshow(valueImage, [])
title('Value Channel', 'FontSize', fontSize);

% Look at the histogram of the hue channel
% so we can see where the blue is
[pixelCounts values] = hist(hueImage, 500);
figure;
subplot(2, 2, 1);
bar(values, pixelCounts);
title('Histogram of Hue Channel', 'FontSize', fontSize);
% Enlarge figure to full screen.
set(gcf, 'units','normalized','outerposition',[0 0 1 1]);
% Blue looks like it's in the 0.4 to 0.7 region.  
% Amplify that by increasing saturation for those pixels.
% Find blue pixels.  They have to have the right hue but not be too bright.
bluePixels = hueImage > 0.4 & hueImage < 0.7 & valueImage < 0.8;
subplot(2, 2, 2);
imshow(bluePixels);
title('Map of Blue Pixels', 'FontSize', fontSize);
% Multiply the saturation channel by 1.5 for those pixels.
saturationImage(bluePixels) = saturationImage(bluePixels) * 3.5;
subplot(2, 2, 3);
imshow(saturationImage);
title('New Saturation Channel', 'FontSize', fontSize);
% Combine back to form new hsv image
hsvImage = cat(3, hueImage, saturationImage, valueImage);
% Convert back to RGB color space.
rgbImage = hsv2rgb(hsvImage);
subplot(2, 2, 4);
imshow(rgbImage);
title('RGB Image with Enhanced Blue', 'FontSize', fontSize);

[Image Analyst]

I'm surprised you didn't think this answer worked. Did you even try it?

[Yicheng]

I like this answer!!!!!!!

[PhD Problems]

@Image Analyst,

Can you please clarify how you selected the blue region from the bar graph?

Here is the generated graph for the sample footbal image used in your cade:

How did you find the blue region to be from 0.4 to 0.7 based on the above? If I wanted to select the region for the brown/orange football, how would I select that color regions from the above graph?

Thank you

[Image Analyst]

Look at my avatar to see the color space. Hues for blue will be around 270 degrees, more or less.

[PhD Problems]

Thank you, that makes sense, but in your code you relied on the above graph and histogram to choose the region of 0.4 to 0.7. I just dont understand how?

As I need to enhance the yellow colour in my sample and based on your profile avatar I can see that that if I want to target yellow, it will be around 0 degrees, right? But how does that transfer to the graph/histogram above?

Thanks

[Image Analyst]

Well I knew that blue was in that range. Yellow is straight up, around 90 degrees. So you can choose a range of about 45 to 135 degrees or so. Divide by 360 to get values in the range 0-1.

Answer 3

[Jan]

RGB = rand(200, 100, 3);
Blue = RGB(:, :, 3);
enhancedBlue = max(Blue + 0.1, 1.0);  % Or whatever
RGB(:, :, 3) = enhancedBlue;