MATLAB: Closing the boundary of an image for segmentation

image processingImage Processing Toolboximage segmentation

Hi everyone,

I am working on segmentation of the particle (see attached). I need to differentiate the environment from the particle interior. I have tried the boundary functions on matlab but with no success. I will certainly appreciate any suggestion on differentiating the outside of the grey areas(environment) from the particle interior. How could I close the particle boundary?

Best Answer

I think combining boundary and poly2mask functions should be some help for your task. How about the following?

% Read the image
I = imread('sampleImg.png');
% Extract non-blue area
BW = I(:,:,3) < 140;
% Apply boundary function
[row,col] = find(BW);
k = boundary(row,col);
% Make mask from boundary result
Mask = poly2mask(col(k),row(k),size(BW,1),size(BW,2));
% Show the result
figure
subplot(1,2,1)
imshow(I)
title('Original Image','FontSize',18)
subplot(1,2,2)
imshow(Mask)
title('Mask','FontSize',18)

Related Solutions

MATLAB: How to find the least pixel intensity value for each connected component

See demo below. Adapt as needed:

grayImage = imread('coins.png');
subplot(2, 2, 1);
imshow(grayImage);
subplot(2, 2, 2);
histogram(grayImage)
binaryImage = imfill(grayImage > 90, 'holes');
binaryImage = bwareafilt(binaryImage, 10);
subplot(2, 2, 3);
imshow(binaryImage);
subplot(2, 2, 4);
[labeledImage, numBlobs] = bwlabel(binaryImage);
props = regionprops(labeledImage, grayImage, 'PixelValues', 'PixelList')
for k = 1 : numBlobs
    % Fin the min value of this blob's pixel values.
    thisBlobsMin = min(props(k).PixelValues);
    % Get locations in this blob
    mask = ismember(labeledImage, k);
	maskedImage = grayImage .* uint8(mask);
    % Find all the rows and columns where it appears
    [rows, columns] = find(maskedImage == thisBlobsMin);
    imshow(maskedImage);
	axis('on', 'image');
	hold on;
	plot(columns, rows, 'r+', 'LineWidth', 2, 'MarkerSize', 15);
	hold off
	caption = sprintf('Showing cross at %d mins for blob #%d of %d.\n', ...
		length(rows), k, numBlobs);
	title(caption, 'FontSize', 12);
	fprintf('%s\n', caption);
end
fprintf('Done!\n');

MATLAB: Remove parts from Image and Fill Parts

Pretty easy. Threshold, fill holes, take the 3 largest.

Solution below. Adapt as needed.

% Initialization steps.
clc;    % Clear the command window.
close all;  % Close all figures (except those of imtool.)
clear;  % Erase all existing variables. Or clearvars if you want.
workspace;  % Make sure the workspace panel is showing.
format long g;
format compact;
fontSize = 14;
filename = fullfile(pwd, 'image_1.jpg');
grayImage = imread(filename);
if ndims(grayImage) == 3
	% It's really an RGB image, not a gray scale image.  
	% Convert to gray scale
	grayImage = rgb2gray(grayImage);
end
subplot(2, 3, 1);
imshow(grayImage, []);
title('Original Image', 'FontSize', fontSize);
% Threshold
subplot(2, 3, 2);
imhist(grayImage);
grid on;
title('Histogram', 'FontSize', fontSize);
mask = grayImage < 128; % or whatever value works
subplot(2, 3, 3);
imshow(mask, []);
title('Initial Binary Image', 'FontSize', fontSize);
% Fill Holes
mask = imfill(mask, 'holes');
subplot(2, 3, 4);
imshow(mask, []);
title('After holes filled', 'FontSize', fontSize);
% Take 3 largest blobs then invert intensity.
mask = ~bwareafilt(mask, 3);
subplot(2, 3, 5);
imshow(mask, []);
title('Final Image Showing 3 Largest', 'FontSize', fontSize);

Best Answer

Related Solutions

MATLAB: How to find the least pixel intensity value for each connected component

MATLAB: Remove parts from Image and Fill Parts

Related Question