I have an OCT images dataset with white border for image acquiring, and I want to delete this white border, I tried imclearborder, but the inner border of the image also deleted , the image below is as the result of imcomlement

[Mrutyunjaya Hiremath]

Plaese , Share some sample input and expected images for better understanding.

[eng had]

the image I attached is the input image , and I need to remove those black padding

[Mrutyunjaya Hiremath]

Use imcrop function where top left coordinate (X1,Y1) and bottom right (X2,Y2), fit for almost all your images

im2 = imcrop(im,[x1 y1 x2 y2]);

Refer here

This is the best and easy way.

[eng had]

I need an automatic way, for the whole dataset

[Mrutyunjaya Hiremath]

can you share 3 to 4 more images to study

Answer 1

[Image Analyst]

I work with OCT images every week. Why do you need to remove them? It shouldn't matter. Anyway, what does remove mean? Like crop? Or set to white? Or set to the mean of the image?

Do you mean like this:

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 = 22;

%--------------------------------------------------------------------------------------------------------
%    READ IN IMAGE
folder = pwd;
baseFileName = 'image.jpeg';
% Get the full filename, with path prepended.
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
% Read in the image from disk.  If storedColorMap is not empty, it's an indexed image with a stored colormap.
[grayImage, storedColorMap] = imread(fullFileName);
if ~isempty(storedColorMap)
	grayImage = ind2rgb(grayImage, storedColorMap);
end
% 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.
	% Use weighted sum of ALL channels to create a gray scale image.
	grayImage = rgb2gray(grayImage);
	% ALTERNATE METHOD: Convert it to gray scale by taking only the green channel,
	% which in a typical snapshot will be the least noisy channel.
	% grayImage = grayImage(:, :, 2); % Take green channel.
end
% Display the image.
hFig = figure;
subplot(2, 2, 1);
imshow(grayImage, []);
impixelinfo;
title('Original Grayscale Image', 'FontSize', fontSize, 'Interpreter', 'None');
impixelinfo;
hFig.WindowState = 'maximized'; % May not work in earlier versions of MATLAB.
drawnow;

%--------------------------------------------------------------------------------------------------------
%    SEGMENTATION
% Binarize the image
binaryImage = grayImage <= 5;
% Take the blobs largest than 20 pixels only.
binaryImage = bwareafilt(binaryImage, [20, inf]);
% Display the segmented image.
subplot(2, 2, 2);
imshow(binaryImage, []);
title('Binary Image', 'FontSize', fontSize, 'Interpreter', 'None');
axis('on', 'image');

%--------------------------------------------------------------------------------------------------------
%    EXTRACTION OF EACH BLOB
props = regionprops(binaryImage, 'BoundingBox', 'Area');
allAreas = [props.Area]
% Display them
for k = 1 : length(props)
	thisBB = props(k).BoundingBox;
	% Erase image in the extended box.
	row1 = ceil(thisBB(2));
	row2 = row1 + thisBB(4);
	col1 = ceil(thisBB(1));
	col2 = col1 + thisBB(3);
	
	grayImage(row1:row2, :) = 0;	
end
% Crop
[r, c] = find(grayImage);
grayImage = grayImage(min(r):max(r), :);
% Display the segmented image.
subplot(2, 2, 3:4);
imshow(grayImage, []);
title('Binary Image', 'FontSize', fontSize, 'Interpreter', 'None');
axis('on', 'image');

[eng had]

Because some image has a white padding larger than the ertina layer, and I'm working of the largerst blob of the image

[eng had]

[Image Analyst]

I don't know what "the ertina layer" is. But anyway, do you have any more questions?

[eng had]

Sorry for the wrong dectation, I meant retina layers... thank for your try. but it 's not working for me

[Image Analyst]

You're a man of very few words. It looks like an image was produced. What's not working? Post your new image and show what cropped image you'd like to obtain as a result.

[eng had]

this is the way, I want to have." No Padding automatically"

[Image Analyst]

OK, we're getting there. Now, what would you like as the result? Do you want to crop off the top triangle, and the left triangle? Will there ever be any triangles on the bottom or right side? Would you want those cropped off too? But you should NEVER use jpeg images for image analysis. The black triangles are blurred due to jpeg artifacts. Please post the image as a PNG image, which will not have any compression artifacts.

[eng had]

the format of the images in the dataset which I downloaded are jpg.

the second image I attached is the result that I need, but if we can crop off all those black triangles, that will be an amazing result.

ps: the orginal images in the dataset have white padding, and I use "imcomplement" to the above image to clarify paddig (surrounding borders)

thank you very much for your response

[Image Analyst]

This will do it:

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 = 22;

%--------------------------------------------------------------------------------------------------------
%    READ IN IMAGE
folder = pwd;
baseFileName = 'image.jpeg';
% Get the full filename, with path prepended.
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
% Read in the image from disk.  If storedColorMap is not empty, it's an indexed image with a stored colormap.
[grayImage, storedColorMap] = imread(fullFileName);
if ~isempty(storedColorMap)
	grayImage = ind2rgb(grayImage, storedColorMap);
end
% 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.
	% Use weighted sum of ALL channels to create a gray scale image.
	grayImage = rgb2gray(grayImage);
	% ALTERNATE METHOD: Convert it to gray scale by taking only the green channel,
	% which in a typical snapshot will be the least noisy channel.
	% grayImage = grayImage(:, :, 2); % Take green channel.
end
% Display the image.
hFig = figure;
subplot(2, 1, 1);
imshow(grayImage, []);
impixelinfo;
title('Original Grayscale Image', 'FontSize', fontSize, 'Interpreter', 'None');
impixelinfo;
hFig.WindowState = 'maximized'; % May not work in earlier versions of MATLAB.
drawnow;

%--------------------------------------------------------------------------------------------------------
%    EXTRACTION OF IMAGE
threshold = 10; % Whatever....
binaryImage = grayImage < threshold;
% Display the segmented image of the triangles.
subplot(2, 1, 2);
imshow(binaryImage, []);
title('Binary Image', 'FontSize', fontSize, 'Interpreter', 'None');
axis('on', 'image');

% Find first non-dark value in each column.
% Extract the parts that overlap the top left corner. But not quite the top right because the top right is not above the threshold, so one row below that..
upperRightTriangle = bwselect(binaryImage, columns, 2);
topRows = ones(columns, 1);
for col = 1 : columns
	t = find(upperRightTriangle(:, col), 1, 'last');
	if ~isempty(t)
		topRows(col) = t;
	end
end
newTopRow = max(topRows)

% Find first non-dark value in each row.
% Extract the parts that overlap the bottom right corner.
lowerLeftTriangle = bwselect(binaryImage, 1, rows-1); % Not quite the bottom because the bottom left is not above the threshold, so one row above that..
leftColumns = ones(rows, 1);
for row = 1 : rows
	t = find(lowerLeftTriangle(row, :), 1, 'last');
	if ~isempty(t)
		leftColumns(row) = t;
	end
end
newLeftColumn = max(leftColumns)

% Crop
croppedGrayImage = grayImage(newTopRow : end, newLeftColumn : end);
% Display the cropped image.
subplot(2, 1, 2);
imshow(croppedGrayImage, []);
title('Cropped Image', 'FontSize', fontSize, 'Interpreter', 'None');
axis('on', 'image');

[eng had]

I tried the new code on my dataset, the code works perfectly fo some image, while the majority of the images look like the attached one

[Image Analyst]

That's because it's been rotated in the opposite direction. You can tell which direction it's been rotated by looking at the width and height of the black triangle in the lower left corner. You can use regionprops() to get the bounding box of the triangles.

So make up two functions, one to crop for a counter-clockwise rotation like I already did for you, and then another, slightly modified function to handle rotation in the other direction. Something like

if heightOfTriangle > widthOfTriangle
    croppedImage = RotateCounterClockwise(grayImage);
else
    croppedImage = RotateClockwise(grayImage);
end

The functions will be largely the same just with slight modifications that I'm sure you can figure out.