how can you determine the threshold that covers the functional area of the image in matlab?

I have an image of a tissue and I want to make a mask of that tissue that is only of the functional area in the image (minus the long black tubular shapes which are blood vessels that I don't want to include in my measurements). How can I make this mask? Here's what i tried:

% Load image
img = imread('tissueimage.tif');
% Convert to grayscale
gray_img = rgb2gray(img);
% Threshold the image to create a binary mask
threshold = 8;
mask = gray_img > threshold;
% Remove small objects
min_size = 1000;
mask = bwareaopen(mask, min_size);
% Fill holes
mask = imfill(mask, 'holes');
% Erode and dilate the mask to remove noise and smooth edges
se = strel('disk', 5);
mask = imerode(mask, se);
mask = imdilate(mask, se);
% Display the original image and the resulting mask
figure;
subplot(1,2,1); imshow(img); title('Original Image');
subplot(1,2,2); imshow(mask); title('Functional Mask');

Here's what I have done so far:

1. Preprocess the image
2. Segment the tissue (removed background)

here's is where I'm unsure

Remove the blood vessels

i did try thresholding and picking numbers from 0-10 but 0, gets too many of the blood vessels and 10, I'm not sure if its all of the functional area or just the white pixels.

here's how i imagine it should look like (masked image). How can i get this?

Thanks pictures attached.

-L

[Image Analyst]

Please post a screenshot.

[Laura]

@Image Analyst masked image -

original image

[Laura]

@Image Analyst

I am trying to select for all the areas excep the connected black areas.

Answer 1

[Image Analyst]

@Laura just take your existing mask (top image) and invert it

mask = ~mask;

[Laura]

@image analyst When I invert it, will it show me the code to allow me impose the same mask on my other images as well?

[Image Analyst]

No, you just get a new image. It does not show or write code for you. You will have to do that yourself. If you need an example, see the links below.

See my Image Segmentation Tutorial in my File Exchange:

Image Analyst's File Exchange

It's a generic, general purpose demo of how to threshold an image to find blobs, and then measure things about the blobs, and extract certain blobs based on their areas or diameters.

Also, see the FAQ if you want to process a sequence of files:

https://matlab.fandom.com/wiki/FAQ#How_can_I_process_a_sequence_of_files?

[Laura]

@Image Analyst what would I do a new image? My goal is to generate a similar mask but all the code i know either selects the white spots or selects the background, there is nothing that selects for most of the background.

Essentially, I want to select against the elongated black tubes (vessels of a minimum length/ connected black pixels) so that it could be distinguished from the background. is there a mask code for that you could show me?

[Image Analyst]

I don't know what you're asking when you say you want to "selects the white spots or selects the background, there is nothing that selects for most of the background". Yes, you can threshold and get the white spots. Or you can threshold and get the black background. I don't know how you got the (what I call) worm image with the elongated shapes, and what those represents. How does it relate to your original image? Do you want to somehow group white spots that are close together into a single elongated region (which will have both background and white spots in it)? Please clarify.

[Laura]

The worms are blood vessels. They are the only thing in the image that I dont want. I want everything else in the image.

In the first image i am attaching you see the worm shapes are the only things black (top image) and everything else is color, after filtering this is the image (middle), and applying the mask (bottom image). But the masks I have made with my above code, don't look like this (they are grainy). after filtering and binarization how can I make a mask that will just remove the black worm shapes and keep everything else?

[Image Analyst]

You can invert your worm mask and use it to set the pixels in the mask you want to false

mask(~wormMask) = false;

[Laura]

@Image Analyst

That sounds like a good idea. But the thing is, the worm mask is a picture of how I want my masks to look like, it is not the mask itself.

I tried to create a similar mask using the image segmenter tool box then generating a code for that mask but it still doesn't seem to make the worm mask. Is there a way to select for the worms in my images then invert it to get the mask?

[Image Analyst]

OK, forget the worm mask. Let's start with what you want to measure. Your original image has white spots, many of which are in clusters, and large regions of black in between the white spots. You can threshold the white spots and get them alone. Isn't that enough? If you do that you won't have the black regions in there and you can just measure the white spots, like their intensity, size distribution, or whatever you want. So who cares about the black worm-like regions if you have your measurements.

Again, forget the worm mask. What do you want to measure on the part of the image not in the worm regions?

[Laura]

@Image Analyst

So the idea is that in the image, there are lots of information (besides the cells and other round things that i can see). Theoretically, by just removing the things that we know are useless (worms) I'd be measuring the proeprties of everything else. So I want to measure everything else in the image.

I guess how do I know that the white spots are that everything? how do i know I'm not removing information about the components of the image if i just threshold the white spots and measure the properties of the white spots?

[Image Analyst]

If you threshold for just the white spots, how would you have any undesired info from the black worm areas? They won't be selected. And I don't think the black worms overlap any regions where you want info. So why can't you do something like

mask = grayImage > someThresholdValue; % Select only bright spots.
props = regionprops('table', 'mask', grayImage, 'all')

See my Image Segmentation Tutorial in my File Exchange:

Image Analyst's File Exchange

It's a generic, general purpose demo of how to threshold an image to find blobs, and then measure things about the blobs, and extract certain blobs based on their areas or diameters.

[Laura]

@Image Analyst

How about I reword: if I wanted to substract out ONLY the worms in each image. How would I do that?

[Laura]

@Image Analyst almost like photoshopping the objects you dont want out of the image..

[Image Analyst]

Use the worm mask as logical indexes to your gray scale image

grayImage(wormMask) = 0;

[Laura]

@Image Analyst the worm mask doesn't exist, its just a photoshop picture of the mask i want to create.

I found your response in another forum that was pretty good! This will work, but my issue is highlighting the red areas that i don't want which isn't showing in my image. Here is the code:

% Read image as RGB

I = imread('bwimage.png'); %including the pull path is better

% Convert to binary

BW = imbinarize(rgb2gray(I));

% Get rid of white border

BWConv = bwconvhull(~BW);

BW = BW & BWConv;

% get region centers and major axis, and orientation

stats = regionprops('table',BW,'BoundingBox','MajorAxisLength','MinorAxisLength');

% Compute length/width ratio

stats.LenWdRatio = stats.MajorAxisLength ./ stats.MinorAxisLength;

% Set threshold and determine which objects

% are greater than the threshold.

thresh = 2.0;

stats.isGreater = stats.LenWdRatio > thresh;

% Show bounding box for objects that met the threshold

figure()

h = imshow(BW);

axis on

hold on

% Plot red rectangles around accepted objects

arrayfun(@(i)rectangle('Position',stats.BoundingBox(i,:),'EdgeColor','r'), find(stats.isGreater));

% Plot yellow rectangles around rejected objects

arrayfun(@(i)rectangle('Position',stats.BoundingBox(i,:),'EdgeColor','y'), find(~stats.isGreater));

I attached my image. mY ONLY ISSUES is how to put the big black circles in the rejected squares. bUt nomatter which threshold I pick the circles don't like up. thANKS very much @Image Analyst

[Image Analyst]

OK but how would you create the worm mask? Describe your thought process.

Answer 2

[Image Analyst]

@Laura 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 TEST IMAGE
folder = [];
baseFileName = 'laura.jpeg';
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
grayImage = 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(grayImage)
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 = rgb2gray(grayImage);
end

%--------------------------------------------------------------------------------------------------------
% Display the image.
subplot(3, 2, 1);
imshow(grayImage, []);
impixelinfo;
axis('on', 'image');
title('Original Gray Scale Image', 'FontSize', fontSize, 'Interpreter', 'None');
% Update 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(grayImage)
% Maximize window.
g = gcf;
g.WindowState = 'maximized';
drawnow;

%--------------------------------------------------------------------------------------------------------
% Get mask
mask = grayImage > 0;
subplot(3, 2, 2);
imshow(mask);
impixelinfo;
axis('on', 'image');
title('Mask Image', 'FontSize', fontSize, 'Interpreter', 'None');
drawnow;

%--------------------------------------------------------------------------------------------------------
% Get Euclidean distance transform
edtImage = bwdist(mask);
subplot(3, 2, 3);
imshow(edtImage, []);
impixelinfo;
axis('on', 'image');
title('EDT Image', 'FontSize', fontSize, 'Interpreter', 'None');
drawnow;

%--------------------------------------------------------------------------------------------------------
% Threshold the edt image to get the large black regions.
mask2 = edtImage > 8;
% The mask has sharp corners.  OPTIONAL: if you want to smooth them out, blur and threshold.
windowWidth = 9;
mask2 = conv2(double(mask2), ones(windowWidth)/windowWidth^2, 'same') > 0.5;
% Looks like maybe we don't want blobs smaller than 1000 pixels.  Remove those. (OPTIONAL)
mask2 = bwareaopen(mask2, 1000);
%--------------------------------------------------------------------------------------------------------
% Measure the areas of all the blobs.
props = regionprops(mask2, 'Area')
allAreas = sort([props.Area])

subplot(3, 2, 4);
imshow(mask2, []);
impixelinfo;
axis('on', 'image');
title('Mask Image', 'FontSize', fontSize, 'Interpreter', 'None');
drawnow;

%--------------------------------------------------------------------------------------------------------
% Plot the borders of all the blobs in the overlay above the original grayscale image 
% using the coordinates returned by bwboundaries().
% bwboundaries() returns a cell array, where each cell contains the row/column coordinates for an object in the image.
subplot(3, 2, 5);
imshow(grayImage); % Optional : show the original image again.  Or you can leave the binary image showing if you want.
% Here is where we actually get the boundaries for each blob.
boundaries = bwboundaries(mask2);
% boundaries is a cell array - one cell for each blob.
% In each cell is an N-by-2 list of coordinates in a (row, column) format.  Note: NOT (x,y).
% Column 1 is rows, or y.    Column 2 is columns, or x.
numberOfBoundaries = size(boundaries, 1); % Count the boundaries so we can use it in our for loop
% Here is where we actually plot the boundaries of each blob in the overlay.
hold on; % Don't let boundaries blow away the displayed image.
for k = 1 : numberOfBoundaries
    thisBoundary = boundaries{k}; % Get boundary for this specific blob.
    x = thisBoundary(:,2); % Column 2 is the columns, which is x.
    y = thisBoundary(:,1); % Column 1 is the rows, which is y.
    plot(x, y, 'r-', 'LineWidth', 2); % Plot boundary in red.
end
hold off;
caption = sprintf('%d Outlines, from bwboundaries()', numberOfBoundaries); 
fontSize = 15;
title(caption, 'FontSize', fontSize); 
axis('on', 'image'); % Make sure image is not artificially stretched because of screen's aspect ratio.

[Laura]

Here’s what I got with my code it was about 3 lines so I will use yours instead because it’s more detailed and thorough.

[Laura]

@Image Analyst

Why is my result grainer?

[Image Analyst]

Looks like you used a lower threshold.

I wish I knew what you really want. What you've shared so far is an XY problem

https://en.wikipedia.org/wiki/XY_problem

I believe you're asking for something that, in the end, really doesn't matter.

[Laura]

The answer you gave answered me perfectly. Thanks for the link.

Why doesn't it really matter though? Thansk for the clarification. @Image Analyst

[Image Analyst]

Because I think you said you want to measure things about the bright blobs. So you can get a mask of those. It's not needed to get the mask of larger dark areas in order to measure the small white blobs.