can i call my binary image as contour map, if yes then how can i extract contour outline without using any dialation approach

i have two thresholded images. one is thresholded by a single threshold and other is threshoded my multiple threshold. can i call those thresholded (binary) as contour-map. if yes how can i get the coordinates of boundary pixels. i want to get the bourndary but without any dialation operation.

Answer 1

[Image Analyst]

I'm not sure what multi-thresholded means to you but if you used imquantize() or something and have a single gray level for each region than you can create a binary image. See this demo:

% Demo to quantize a grayscale image and find boundaries.
clc;    % Clear the command window.
close all;  % Close all figures (except those of imtool.)
imtool close all;  % Close all imtool figures if you have the Image Processing Toolbox.
clear;  % Erase all existing variables. Or clearvars if you want.
workspace;  % Make sure the workspace panel is showing.
format long g;
format compact;
fontSize = 13;

% Create a sample image.
grayImage = uint8(255*mat2gray(peaks(300)));
fh = figure();
% 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, 'units','normalized','outerposition',[0 0 1 1]);

edges = [50, 100, 150];
quantizedImage = imquantize(grayImage, edges);
% Display the quantized image.
subplot(2, 2, 2);
imshow(quantizedImage, []);
title('Quantized Grayscale Image', 'FontSize', fontSize);

for r = 1 : length(edges)
  % Extract the binary image for region #2
  binaryImage = quantizedImage == r;
  % Display the quantized image.
  subplot(2, 2, 3);
  imshow(binaryImage, []);
  title('Binary Image', 'FontSize', fontSize);

  hold on;
  boundaries = bwboundaries(binaryImage);
  numberOfBoundaries = size(boundaries, 1);
  for k = 1 : numberOfBoundaries
    thisBoundary = boundaries{k};
    x = thisBoundary(:,2);
    y = thisBoundary(:,1);
    plot(x, y, 'r-', 'LineWidth', 4);
  end
  hold off;
  promptMessage = sprintf('Do you want to Continue processing,\nor Cancel to abort processing?');
  titleBarCaption = 'Continue?';
  button = questdlg(promptMessage, titleBarCaption, 'Continue', 'Cancel', 'Continue');
  if strcmpi(button, 'Cancel')
    return;
  end
end
close(fh);

[Bilal]

thresh = multithresh(I,2); seg_I = imquantize(I,thresh);

here the seg_I will be segmented image with 3 regions. i am calling this as multithresholded image.

[Image Analyst]

Great, so my code will work for you.

Answer 2

[Walter Roberson]

bwboundaries(), bwtraceboundary()

[Bilal]

this work on binary image, 0 and 1. what about the multi-thresholded image? these functions treat all non-zero items as single object.