How to identify boundaries of shapes in an image using the change in color

Hi I am new to Matlab, I have got to identify boundaries of irregular shapes in Image using the change in color, Please guide me how to get info from image into a matrix

Answer 1

[Image Analyst]

If you want the color edges, then the most sensible thing is to measure Delta E - the color difference - between a pixel and its neighbors. I just gave the algorithm a few days ago here: http://www.mathworks.com/matlabcentral/answers/73635#answer_83593 An edge will have a high delta E while non-edges will have low delta E. I now have a demo script of this. Let me know if you want me to post it.

[Yash]

Please post the demo script.

[Image Analyst]

OK, here it is. Please "Accept" the answer if it answered your question.

% Demo to computer color contrast using the CIE Color Difference Formula (Delta E).
% Has the effect of finding color edges (high contrast regions),
% and "uniform" regions (low contrast regions).
% The RGB image is converted to LAB color space and then the
% mean color in a localized window is computed (i.e. the local average image).
% The Delta E (the color difference in LAB color space) is then calculated
% for every pixel in the image between that pixel's color and the average
% LAB color of the window region.  The user can then specify a number that
% says how close to that color would they like to be.  The software will
% then find all pixels within that specified Delta E of the color each pixel.
%
function LocalDeltaE()
clc;  % Clear command window.
clear;  % Delete all variables.
close all;  % Close all figure windows except those created by imtool.
% imtool close all;  % Close all figure windows created by imtool.
workspace;  % Make sure the workspace panel is showing.

% Change the current folder to the folder of this m-file.
if(~isdeployed)
  cd(fileparts(which(mfilename)));
end

try
  % Check that user has the Image Processing Toolbox installed.
  hasIPT = license('test', 'image_toolbox');
  if ~hasIPT
    % User does not have the toolbox installed.
    message = sprintf('Sorry, but you do not seem to have the Image Processing Toolbox.\nDo you want to try to continue anyway?');
    reply = questdlg(message, 'Toolbox missing', 'Yes', 'No', 'Yes');
    if strcmpi(reply, 'No')
      % User said No, so exit.
      return;
    end
  end
  
  % Continue with the demo.  Do some initialization stuff.
  close all;
  fontSize = 14;
  figure;
  defaultThresholdAnswer = {'7'};  % Suggest this number to the user to threshold the contrast.
  % Maximize the figure.
  set(gcf, 'Position', get(0, 'ScreenSize'));
  set(gcf,'name','Local Color Contrast Demo','numbertitle','off')
  
  % Change the current folder to the folder of this m-file.
  % (The line of code below is from Brett Shoelson of The Mathworks.)
  if(~isdeployed)
    cd(fileparts(which(mfilename)));
  end
  
  % Ask user if they want to use a demo image or their own image.
  message = sprintf('Do you want use a standard demo image,\nOr pick one of your own?');
  reply2 = questdlg(message, 'Which Image?', 'Demo','My Own', 'Demo');
  % Open an image.
  if strcmpi(reply2, 'Demo')
    % Read standard MATLAB demo image.
    message = sprintf('Which demo image do you want to use?');
    selectedImage = questdlg(message, 'Which Demo Image?', 'Onions', 'Peppers', 'Trees', 'Peppers');
    if strcmp(selectedImage, 'Onions')
      fullImageFileName = 'onion.png';
      defaultThresholdAnswer = {'5'};  % Suggest this number to the user.
    elseif strcmp(selectedImage, 'Peppers')
      fullImageFileName = 'peppers.png';
      defaultThresholdAnswer = {'8'};  % Suggest this number to the user.
    else
      fullImageFileName = 'trees.tif';
      defaultThresholdAnswer = {'20'};  % Suggest this number to the user.
    end
  else
    % They want to pick their own.
    % Change default directory to the one containing the standard demo images for the MATLAB Image Processing Toolbox.
    originalFolder = pwd;
    folder = fullfile(matlabroot, '\toolbox\images\imdemos');
    if ~exist(folder, 'dir')
      folder = pwd;
    end
    cd(folder);
    % Browse for the image file.
    [baseFileName, folder] = uigetfile('*.*', 'Specify an image file');
    fullImageFileName = fullfile(folder, baseFileName);
    % Set current folder back to the original one.
    cd(originalFolder);
  end
  
  % Check to see that the image exists.  (Mainly to check on the demo images.)
  if ~exist(fullImageFileName, 'file')
    message = sprintf('This file does not exist:\n%s', fullImageFileName);
    WarnUser(message);
    return;
  end
  
  % Read in image into an array.
  [rgbImage, storedColorMap] = imread(fullImageFileName);
  [rows, columns, numberOfColorBands] = size(rgbImage);
  % If it's monochrome (indexed), convert it to color.
  % Check to see if it's an 8-bit image needed later for scaling).
  if strcmpi(class(rgbImage), 'uint8')
    % Flag for 256 gray levels.
    eightBit = true;
  else
    eightBit = false;
  end
  if numberOfColorBands == 1
    if isempty(storedColorMap)
      % Just a simple gray level image, not indexed with a stored color map.
      % Create a 3D true color image where we copy the monochrome image into all 3 (R, G, & B) color planes.
      rgbImage = cat(3, rgbImage, rgbImage, rgbImage);
    else
      % It's an indexed image.
      rgbImage = ind2rgb(rgbImage, storedColorMap);
      % ind2rgb() will convert it to double and normalize it to the range 0-1.
      % Convert back to uint8 in the range 0-255, if needed.
      if eightBit
        rgbImage = uint8(255 * rgbImage);
      end
    end
  end
  
  % Display the original image.
  subplot(3, 4, 1);
  imshow(rgbImage);
  drawnow; % Make it display immediately.
  if numberOfColorBands > 1
    title('Original Color Image', 'FontSize', fontSize);
  else
    caption = sprintf('Original Indexed Image\n(converted to true color with its stored colormap)');
    title(caption, 'FontSize', fontSize);
  end
  
  % Convert image from RGB colorspace to lab color space.
  cform = makecform('srgb2lab');
  lab_Image = applycform(im2double(rgbImage),cform);
  
  % Extract out the color bands from the original image
  % into 3 separate 2D arrays, one for each color component.
  LChannel = lab_Image(:, :, 1);
  aChannel = lab_Image(:, :, 2);
  bChannel = lab_Image(:, :, 3);
  
  % Display the lab images.
  subplot(3, 4, 2);
  imshow(LChannel, []);
  title('L Channel (Overall Lightness)', 'FontSize', fontSize);
  subplot(3, 4, 3);
  imshow(aChannel, []);
  title('a Channel (redness to greenness)', 'FontSize', fontSize);
  subplot(3, 4, 4);
  imshow(bChannel, []);
  title('b Channel (yellowness to blueness)', 'FontSize', fontSize);
  
  % Find out the square window size where wants to average the colors.
  prompt = {sprintf('What square window size do you want to take the average color within? ')};
  dialogTitle = 'Enter the window size';
  numberOfLines = 1;
  defaultAnswer = {'5'};  % Default window width.
  response = inputdlg(prompt, dialogTitle, numberOfLines, defaultAnswer);
  % Update tolerance with user's response.
  windowSize = str2double(cell2mat(response));
  
  % Get the average lab color value, localized within that window.
  [LMean, aMean, bMean] = GetMeanLABValues(LChannel, aChannel, bChannel, windowSize);
  
  % Get the elta colors for each LAB color channel.
  deltaL = LMean - LChannel;
  deltaa = aMean - aChannel;
  deltab = bMean - bChannel;
  
  % Create the Delta E image.
  % This is an image that represents the color difference.
  % Delta E is the square root of the sum of the squares of the delta images.
  deltaE = sqrt(deltaL .^ 2 + deltaa .^ 2 + deltab .^ 2);
  
  % Display the masked Delta E image - the delta E within the masked region only.
  subplot(3, 4, 5);
  imshow(deltaE, []);
  caption = sprintf('Delta E between original color image\nand the mean color within the local window region.');
  title(caption, 'FontSize', fontSize);
  
  % Plot the histogram of the Delta E color difference image.
  caption = sprintf('Histogram of the Delta E Image\n');
  PlotHistogram(deltaE, [3 4 6], caption);
  
  message = sprintf('Regions dissilimar in color to the local color\nwill be bright in the Delta E image.\nLow contrast regions will be dark.');
  msgboxw(message);
  
  % Find out how close the user wants to match the colors.
  prompt = {sprintf('First, examine the histogram.\nThen find pixels within this Delta E that you want to threshold at:')};
  dialogTitle = 'Enter Delta E to threshold at';
  numberOfLines = 1;
  % defaultThresholdAnswer was determined when they read in the demo image.
  response = inputdlg(prompt, dialogTitle, numberOfLines, defaultThresholdAnswer);
  % Update tolerance with user's response.
  thresholdValue = str2double(cell2mat(response));
  
  % Let them interactively select the threshold with the threshold() m-file.
  % (Note: This is a separate function in a separate file in my File Exchange.)
  % http://www.mathworks.com/matlabcentral/fileexchange/29372-thresholding-an-image
  %   threshold(deltaE);
  
  % Place a vertical bar at the threshold location.
  handleToSubPlot8 = subplot(3, 4, 6);  % Get the handle to the plot.
  PlaceVerticalBarOnPlot(handleToSubPlot8, thresholdValue, [0 .5 0]);  % Put a vertical red line there.
  
  % Find pixels within that delta E.
  binaryImage = deltaE >= thresholdValue;
  % Display high contrast mask.
  subplot(3, 4, 7);
  imshow(binaryImage, []);
  caption = sprintf('Mask of High Delta Es (Delta E >= %.1f)', thresholdValue);
  title(caption, 'FontSize', fontSize);
  % Display low contrast mask.
  subplot(3, 4, 8);
  imshow(~binaryImage, []);
  caption = sprintf('Mask of Low Delta Es (Delta E <= %.1f)', thresholdValue);
  title(caption, 'FontSize', fontSize);
  
  % Mask the image with the high contrast colors and extract those pixels.
  highContrastPixelImage = bsxfun(@times, rgbImage, cast(binaryImage, class(rgbImage)));
  subplot(3, 4, 11);
  imshow(highContrastPixelImage);
  caption = sprintf('High Contrast Pixel Colors (Delta E <= %.1f)', thresholdValue);
  title(caption, 'FontSize', fontSize);
  
  % Mask the image with the low contast colors and extract those pixels.
  lowContrastPixelsImage = bsxfun(@times, rgbImage, cast(~binaryImage, class(rgbImage)));
  subplot(3, 4, 12);
  imshow(lowContrastPixelsImage);
  caption = sprintf('Low Contrast Pixel Colors (Delta E > %.1f)', thresholdValue);
  title(caption, 'FontSize', fontSize);
  
  % Display credits: the MATLAB logo and my name.
  ShowCredits(); % Display logo in plot position #12.
  
  % Alert user that the demo has finished.
  message = sprintf('Done!\n\nThe demo has finished.');
  msgboxw(message);
  
catch ME
  errorMessage = sprintf('Error running this m-file:\n%s\n\nThe error message is:\n%s', ...
    mfilename('fullpath'), ME.message);
  errordlg(errorMessage);
end
return; % from SimpleColorDetection()
% ---------- End of main function ---------------------------------

%----------------------------------------------------------------------------
% Display the MATLAB logo.
function ShowCredits()
try
  logoFig = subplot(3, 4, 9);
  caption = sprintf('A MATLAB Demo\nby Image Analyst');
  text(0.5,1.15, caption, 'Color','r', 'FontSize', 18, 'FontWeight','b', 'HorizontalAlignment', 'Center') ;
  positionOfLowerRightPlot = get(logoFig, 'position');
  L = 40*membrane(1,25);
  logoax = axes('CameraPosition', [-193.4013 -265.1546  220.4819],...
    'CameraTarget',[26 26 10], ...
    'CameraUpVector',[0 0 1], ...
    'CameraViewAngle',9.5, ...
    'DataAspectRatio', [1 1 .9],...
    'Position', positionOfLowerRightPlot, ...
    'Visible','off', ...
    'XLim',[1 51], ...
    'YLim',[1 51], ...
    'ZLim',[-13 40], ...
    'parent',gcf);
  s = surface(L, ...
    'EdgeColor','none', ...
    'FaceColor',[0.9 0.2 0.2], ...
    'FaceLighting','phong', ...
    'AmbientStrength',0.3, ...
    'DiffuseStrength',0.6, ...
    'Clipping','off',...
    'BackFaceLighting','lit', ...
    'SpecularStrength',1.1, ...
    'SpecularColorReflectance',1, ...
    'SpecularExponent',7, ...
    'Tag','TheMathWorksLogo', ...
    'parent',logoax);
  l1 = light('Position',[40 100 20], ...
    'Style','local', ...
    'Color',[0 0.8 0.8], ...
    'parent',logoax);
  l2 = light('Position',[.5 -1 .4], ...
    'Color',[0.8 0.8 0], ...
    'parent',logoax);
  
catch ME
  errorMessage = sprintf('Error running ShowCredits().\n\nThe error message is:\n%s', ...
    ME.message);
  errordlg(errorMessage);
end
return; % from ShowCredits()

%-----------------------------------------------------------------------------
% Get the average lab color within the window region.
% Get new images where each pixel of the output image is the average
% of the pixels in a square window, of width windowSize,
% around the input pixel (including the central pixel).
function [LMean, aMean, bMean] = GetMeanLABValues(LChannel, aChannel, bChannel, windowSize)
try
  kernel = ones(windowSize, windowSize) / windowSize^2;
  LMean = conv2(LChannel, kernel, 'same'); % Average of only the pixels within the windowed area.
  aMean = conv2(aChannel, kernel, 'same'); % Average of only the pixels within the windowed area.
  bMean = conv2(bChannel, kernel, 'same'); % Average of only the pixels within the windowed area.
catch ME
  errorMessage = sprintf('Error running GetMeanLABValues:\n\n\nThe error message is:\n%s', ...
    ME.message);
  WarnUser(errorMessage);
end
return; % from GetMeanLABValues

%==========================================================================================================================
function WarnUser(warningMessage)
uiwait(warndlg(warningMessage));
return; % from WarnUser()

%==========================================================================================================================
function msgboxw(message)
uiwait(msgbox(message));
return; % from msgboxw()

%==========================================================================================================================
% Plots the histograms of the pixels in both the masked region and the entire image.
function PlotHistogram(doubleImage, plotNumber, caption)
try
  fontSize = 14;
  numberOfBins = 50;
  subplot(plotNumber(1), plotNumber(2), plotNumber(3));
  
  % Find out where the edges of the histogram bins should be.
  maxValue = max(doubleImage(:));
  histBinEdges = linspace(0, maxValue, numberOfBins);
  
  % Get the histogram of the entire image into 100 bins.
  pixelCount = histc(doubleImage(:), histBinEdges);
  
  % Plot the  histogram of the entire image.
  bar(histBinEdges, pixelCount, 'BarWidth', 1);
  title(caption, 'FontSize', fontSize);
  
  % Scale x axis manually.
  xlim([0 histBinEdges(end)]);
catch ME
  errorMessage = sprintf('Error running PlotHistogram:\n\n\nThe error message is:\n%s', ...
    ME.message);
  WarnUser(errorMessage);
end
return; % from PlotHistogram

%=====================================================================
% Shows vertical lines going up from the X axis to the curve on the plot.
function lineHandle = PlaceVerticalBarOnPlot(handleToPlot, x, lineColor)
try
  % If the plot is visible, plot the line.
  if get(handleToPlot, 'visible')
    axes(handleToPlot);  % makes existing axes handles.axesPlot the current axes.
    % Make sure x location is in the valid range along the horizontal X axis.
    XRange = get(handleToPlot, 'XLim');
    maxXValue = XRange(2);
    if x > maxXValue
      x = maxXValue;
    end
    % Erase the old line.
    %hOldBar=findobj('type', 'hggroup');
    %delete(hOldBar);
    % Draw a vertical line at the X location.
    hold on;
    yLimits = ylim;
    lineHandle = line([x x], [yLimits(1) yLimits(2)], 'Color', lineColor, 'LineWidth', 3);
    hold off;
  end
catch ME
  errorMessage = sprintf('Error running PlaceVerticalBarOnPlot:\n\n\nThe error message is:\n%s', ...
    ME.message);
  WarnUser(errorMessage);
end
return;  % End of PlaceVerticalBarOnPlot

[Valeska Pearson]

You use the GetMeanLABValues function here in the code, where can I find the this function's code?

[Image Analyst]

It's right in there. Always has been. Look again.

[ankita mandowara]

GetMeanLabValues is giving an error as--- function unidentified...what to do?

[Image Analyst]

Copy and paste the error back here. Paste ALL the red text so I can see what went wrong. The only functions in there are ones() and conv2(), and those are built-in functions of base MATLAB so I think it should be okay.

[ankita taparia]

I am pasting small small portion of code in matlab and then running it. Do you suggest to execute complete code all together?

[Image Analyst]

You didn't paste anything.

Obviously if you just execute some partial chunk of the code, it may not run correctly.

[ankita taparia]

Hey, have executed complete code successfully on my image but couldn't understand a few results. Would you please have a look at the kind of image i am working on?

[Image Analyst]

OK. You should have attached it to your comment, and I would have looked at it, but unfortunately you forgot to insert it. Now I'm going to shut down for the day. I'll look at it tomorrow. Also say what results you couldn't understand with my code.

[ankita taparia]

Sorry for the delay,but please help me. I am attaching my image.

[ankita taparia]

I need to save each individual region separately.

[Image Analyst]

You might try a filter, like

DEM

or a mean shift filter, or a Kuwahara filter.

[ankita taparia]

Mean shift filter is not giving satisfactory result. And DEM code is giving errors which I am not able to fix. Please help.

[ankita taparia]

Also, I tried Sobel filter and getting decent results with it..only if i could be suggested one or two ilters like sobel(not canny) so that i could compare and proceed further.

Answer 2

[Ahmed A. Selman]

There are few files in the file-exchange http://www.mathworks.com/matlabcentral/fileexchange/ so do some search there, you might find what you need.

Or, you can convert the colored image to binary one, then use the function (edge).

[Yash]

I have searched but couldn't find useful sol

[Ahmed A. Selman]

See this file:

http://www.mathworks.com/matlabcentral/fileexchange/28114-fast-edges-of-a-color-image-actual-color-not-converting-to-grayscale

Answer 3

[Yash]

Thanks for the script, I think this will definitely be helpful to me. I need to construct a geometry through this So I need to get the Coordinates of the boundaries created in the image as well as the colour code to know about the orientation. Please help me how can I convert the output that I have got to a matrix containing the required information.

Regards yash