How to detect rectangle in an image then crop it out?

Hi All,

I wish to detect the rectangle as shown below in the image. After that, I want to crop out what's only inside the rectangle part.

Anyone can teach me how?

First, I changed this coloured image into grayscale image then to binary image. Then, I tried using 'regionprops' with 'BoundingBox' to detect the presence of rectangle. However, the cropped image is not exactly the rectangle size.

PS. I do refer from someone else code.

% Find boundary box 
props = regionprops(binaryImage, 'BoundingBox');
boundingBox = props.BoundingBox;
width = boundingBox(3);
height = boundingBox(4);
% Crop off the top, shaded portion of column headers,
% which are known to be a fixed 29% of the height of the box.
newHeight = height * (1 - 0.29)
% Make a new bounding box.
xLeft = boundingBox(1);
yTop = boundingBox(2);
boundingBox2 = [xLeft, yTop + 0.29 * height, width, newHeight];
% Crop the image
newGrayScaleImage = imcrop(grayImage, boundingBox2);
subplot(2, 3, 6);
imshow(newGrayScaleImage);
axis('on', 'image');
impixelinfo;
title('New Gray Scale Image', 'FontSize', fontSize);

Result:

Answer 1

[Akira Agata]

How about the following?

% Load the image and convert it to gray scale
I = imread('image.jpeg');
Igray = rgb2gray(I);

% Detect black line
BW = Igray < 10;

% Detect rectangle region (inside the line)
BW = imclearborder(~BW);
BW = bwareafilt(BW,1);

% Calculate bounding box
s = regionprops(BW,'BoundingBox');

% Crop the image
Icropped = imcrop(I,s.BoundingBox);

% Show the result
figure
imshow(Icropped)

[WanYu]

Hi,

Thank you for answering my question. I am currently using this method now.

However, if I wish to detect another shape on the image as shown in the figure below, how can I achieve that?

I wish to detect the black outline in the image and crop it out later.

Thank you if you are willing to help. :)

[Akira Agata]

I believe the same strategy will be applicable. How about the following?

% Load the image and convert it to gray scale
I = imread('image.jpeg');
Igray = rgb2gray(I);

% Detect black line
BW = Igray < 10;

% Detect the region (inside the line)
BW = imclearborder(~BW);
BW = bwareafilt(BW,1);

% Fill the outside region with black color
I = immultiply(I,repmat(BW,[1 1 3])); % <- Only added this process

% Calculate bounding box
s = regionprops(BW,'BoundingBox');

% Crop the image
Icropped = imcrop(I,s.BoundingBox);

% Show the result
figure
imshow(Icropped)

[WanYu]

Hi,

Thank you for the reply. It really helps me a lot!

[WanYu]

Hi,

Can you please explain to me, why the repmat is using (1,1,3)?

[Akira Agata]

This is because the array I is N x M x 3 (RGB) matrix. To apply immultiply to I, the N x M binary array BW should be expanded to N x M x 3. So I have applied repmat using [1,1,3].