Morphological Method to Remove Tails of Ellipse Shape Sperms

Hi all;

I have used Otsu Thresholding in order to get the attached image (sperm Images). I want to segment the sperm heads. The problem is about the tails depicted in the image. I have some ellipse-like sperm heads that sometimes lines (or other extra parts) are attached to them. I need a morphological method (or any other methods) to remove the extra parts from binary image so that just "ellipse" or "circle"-like objects remain. In other words "Is there any morphological process that is sensitive to the shape of the objects?".

Is there any way to do this, without changing the ellipse-like heads of the sperms?

Thank you very much;

Peyman

Answer 1

[Image Analyst]

Try this:

% Demo by Image Analyst
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 = 16;
markerSize = 40;
%--------------------------------------------------------------------------------------------------------
%    READ IN IMAGE
folder = pwd;
baseFileName = 'sperm.jpeg';
% baseFileName = 'org.jpg';
fullFileName = fullfile(folder, baseFileName);
% Check if file exists.
if ~exist(fullFileName, 'file')
    % The file doesn't exist -- didn't find it there in that folder.
    % Check the entire search path (other folders) for the file by stripping off the folder.
    fullFileNameOnSearchPath = baseFileName; % No path this time.
    if ~exist(fullFileNameOnSearchPath, 'file')
        % Still didn't find it.  Alert user.
        errorMessage = sprintf('Error: %s does not exist in the search path folders.', fullFileName);
        uiwait(warndlg(errorMessage));
        return;
    end
end
rgbImage = imread(fullFileName);
% Get the dimensions of the image.
% numberOfColorChannels should be = 1 for a gray scale image, and 3 for an RGB color image.
[rows, columns, numberOfColorChannels] = size(rgbImage)
if numberOfColorChannels > 1
    % It's not really gray scale like we expected - it's color.
    fprintf('It is not really gray scale like we expected - it is color\n');
    % Extract the blue channel.
    grayImage = rgbImage(:, :, 1);
else
    grayImage = rgbImage;
end
%--------------------------------------------------------------------------------------------------------
% Display the image.
subplot(2, 3, 1);
imshow(grayImage);
impixelinfo;
axis('on', 'image');
title('Original Gray Scale Image', 'FontSize', fontSize, 'Interpreter', 'None');
% Maximize window.
g = gcf;
g.WindowState = 'maximized';
drawnow;
%--------------------------------------------------------------------------------------------------------
% Display histogram
subplot(2, 3, 2);
histogram(grayImage, 256);
grid on;
title('Histogram of Image', 'FontSize', fontSize, 'Interpreter', 'None');
%--------------------------------------------------------------------------------------------------------
% Binarize the image to get a mask.
threshold = 128;
mask = grayImage < threshold;
% Display mask image.
subplot(2, 3, 3);
imshow(mask);
axis('on', 'image');
drawnow;
caption = sprintf('Mask using a threshold of %.1f', threshold);
title(caption, 'FontSize', fontSize, 'Interpreter', 'None');
%--------------------------------------------------------------------------------------------------------
% Do a morphological opening on the image.
radius = 3;  % You can adjust this parameter to affect how much gets removed.
se = strel('disk', radius, 0);
mask = imopen(mask, se);
% Get rid of abnormally small blobs (noise) and unreadlistically large blobs.
% Only take blobs between 10 pixels and 2000 pixels.
mask = bwareafilt(mask, [10, 2000]);
subplot(2, 3, 4);
imshow(mask);
axis('on', 'image');
drawnow;
title('Opened Image', 'FontSize', fontSize, 'Interpreter', 'None');
%--------------------------------------------------------------------------------------------------------
% Label each blob with 8-connectivity, so we can make measurements of it
[labeledImage, numberOfBlobs] = bwlabel(mask, 8);
% Apply a variety of pseudo-colors to the regions.
coloredLabelsImage = label2rgb (labeledImage, 'hsv', 'k', 'shuffle');
% Display the pseudo-colored image.
subplot(2, 3, 5);
imshow(coloredLabelsImage);
% Get all the blob properties.
blobMeasurements = regionprops(labeledImage, 'all');
numberOfBlobs = size(blobMeasurements, 1);
caption = sprintf('%d Individual Sperms', numberOfBlobs)
title(caption, 'FontSize', fontSize, 'Interpreter', 'None');
% Get the areas
allAreas = [blobMeasurements.Area];
% Get the diameters
allDiameters = [blobMeasurements.EquivDiameter];
% Show size histogram of Areas.
subplot(2, 3, 6);
histogram(allAreas);
grid on;
xlabel('Area in Pixels', 'FontSize', fontSize, 'Interpreter', 'None');
ylabel('Count', 'FontSize', fontSize, 'Interpreter', 'None');
title('Histogram of Areas', 'FontSize', fontSize, 'Interpreter', 'None');

[Javane Javaherchian]

Thanks a lot. It is really helpful. I have another question. If I want to track the blob in video, how can I draw the bounderies around the moving blob instead of circle or rectangle? I mean I know the bwbounderies() for image. Is there any similar functuon for video or can I use this function for video?how? I just need the best fit of ellipse or bounderies around the moving blob for tracking. Do you have any idea?

[Image Analyst]

No, you have to do it a frame at a time. Then you can use the actual boundaries from bwboundaries, or the bounding box or ellipse from regionprops.

[Javane Javaherchian]

Okay, thanks a lot.

Answer 2

[Javane Javaherchian]

Hi Peyman,

I have the same problem. Would you please tell that you could find the answer?

Many thanks,

Javane

Answer 3

[Image Analyst]

Did you try imclose?

[Javane Javaherchian]

Hi,

Yes. But it did not work. I want to remove the part of object (tail here) and just thresholding the head in video. Do you have any other suggestion? Many thanks