how to remove digits and axis marks from ultra sound images, which are displayed on the image

Hey everyone,

Im trying to pre-process ultra sound images. I have successfuly extracted only the Ultra Sound image itself from the medical imaging of the machine (removed all additinal unneccesary information around).

Now, I have a problem that for several images there are digits and axis marks on the left side of the extracted image.

Is there any way to remove those marks without just puting 0 on them? Im trying to make average of their neighboors, so I would lose as less information as possible.

Attaching the image, with specified marked in red.

Thanks!

[Walter Roberson]

Detect them and replace them with nan and then use File Exchange inpaint_nan by John D'Errico?

[Itzhak Mamistvalov]

sorry for the delay, I have missed your answer. However, looks like this file fits my goals, but how should i find thos like arrow-shaped shapes in the image? (Im fine with the numbers and their scale)

[DGM]

You might want to include example images. In fact, these look like screenshots. Does the source imaging/viewer software not allow for export of cleaner images? I imagine there's an underlying DICOM archive and none of this is necessary.

[Itzhak Mamistvalov]

Well, I guess there might be an option. However, Im actually using this data-base which includes these style of images (I dont have an option to re-take the images from the software).

[DGM]

Attach an example file without the red annotations.

[Itzhak Mamistvalov]

This is an example image. I have successfully eliminated the numbers and their scales, but cant actualy take off the like-arrow shapes.

My purpose is to get rid off all unnecessary data and stay with the ultrasound image itself (the relevant data, in the center).

Answer 1

[Image Analyst]

Since the marks will be rougly in the same locations, you can use a predetermined mask to isolate just those regions, then use regionfill(). Untested code:

mask1 = false(size(grayImage));
mask1(:, col1:col2) = true; % etc., using locations you know.
% Get gray image in mask region
maskedImage = grayImage;
maskedImage(~mask1) = 0; % So we won't find white occurring in the middle of the image.
% Find pure white in those regions
mask2 = maskedImage == 255;
% use regionfill() to fill them in
repairedImage = reginofill(grayImage, mask2);

Adapt as needed, especially in where you define regions for mask1.

Answer 2

[Image Analyst]

@Itzhak Mamistvalov, 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 = 20;

%--------------------------------------------------------------------------------------------------------
%    READ IN IMAGE
folder = pwd;
baseFileName = 'ultrasound.png';
grayImage = imread(baseFileName);

% 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.
	% Extract the red channel (so the magenta lines will be white).
	grayImage = grayImage(:, :, 1);
end

%--------------------------------------------------------------------------------------------------------
% Display the image.
subplot(2, 3, 1);
imshow(grayImage, []);
axis('on', 'image');
title('Original Image', 'FontSize', fontSize, 'Interpreter', 'None');
impixelinfo;
hFig = gcf;
hFig.WindowState = 'maximized'; % May not work in earlier versions of MATLAB.
drawnow;

% Crop image using known locations.
grayImage = grayImage(226:849, 1:837);
% Display the image.
subplot(2, 3, 2);
imshow(grayImage, []);
axis('on', 'image');
title('Cropped Image', 'FontSize', fontSize, 'Interpreter', 'None');
impixelinfo;


% Binarize
mask1 = false(size(grayImage));
mask1(:, 1:165) = true;
mask1(:, 743:end) = true;
% Get gray image in mask region
% Display the image.
subplot(2, 3, 3);
imshow(mask1, []);
axis('on', 'image');
title('Mask 1', 'FontSize', fontSize, 'Interpreter', 'None');
impixelinfo;

maskedImage = grayImage;
maskedImage(~mask1) = 0; % So we won't find white occurring in the middle of the image.
% Find pure white in those regions
mask2 = maskedImage >= 146; % Whatever...
% Don't mask anything inside the main part of the image.
mask2 = mask2 & mask1;
% Dilate it a little
mask2 = imdilate(mask2, true(3));
% Display the image.
subplot(2, 3, 4);
imshow(mask2, []);
axis('on', 'image');
title('Mask 2', 'FontSize', fontSize, 'Interpreter', 'None');
impixelinfo;

% use regionfill() to fill them in
repairedImage = regionfill(grayImage, mask2);

subplot(2, 3, 5);
imshow(repairedImage, []);
axis('on', 'image');
title('Repaired Image', 'FontSize', fontSize, 'Interpreter', 'None');
impixelinfo;
drawnow;