MATLAB Answers

How to classify shapes of this image as square, rectangle, triangle and circle?

289 views (last 30 days)
ROMIL
ROMIL on 19 Feb 2014
Commented: Dina Abd El-twab on 24 Feb 2020
Please provide me the matlab code to identify shapes on this image and classify them as square, rectangle, circle and triangle.

  5 Comments

Show 2 older comments
phaneendra ch
phaneendra ch on 11 Dec 2015
Your code is working very excellent but I am getting a problem I.e., if there is any small hole in the object the code is unable to generate matrix for it (or getting an error while exexcution). How can I fill up small enclosed holes in black and white image? And can I get a different subplots for different shapes(or different objects) in a images?
Image Analyst
Image Analyst on 12 Dec 2015
Not sure whose code you're talking about, but glad you finally got it working. If you have any questions in the future, post your image and code in a new question.

Sign in to comment.

Accepted Answer

Matt Kindig
Matt Kindig on 21 Feb 2014
Edited: Matt Kindig on 21 Feb 2014
Another approach is to calculate the best-fit bounding rectangle of each object, such that the bounding rectangle can be oriented at an arbitrary angle. I took the following approach:
1) Identify the boundary (i.e. perimeter) of each object, using bwboundaries()
2) Calculate the smallest rectangular bounding box that contains this perimeter. To do this, I used the minboundrect function available at http://www.mathworks.com/matlabcentral/fileexchange/34767-a-suite-of-minimal-bounding-objects/content/MinBoundSuite/minboundrect.m.
3) I calculated the width, height, and area of each bounding rectangle. The aspect ratio (ratio between the width and height) can be used to determine whether it is a square (aspect ratio ~= 1.0) or rectangle.
4) For a rectangle or square, the filled area of the object (from regionprops()) should be almost the same as the area of its bounding rectangle, whereas for a triangle it should be substantially less.
5) For the circularity condition, I used the ratio between perimeter and area like Image Analyst suggested.
Enjoy!
im = imread('http://www.mathworks.com/matlabcentral/answers/uploaded_files/8372/abc.jpg');
%convert to 2D black and white with colors inverted
BW = im(:,:,1) < 10;
%get outlines of each object
[B,L,N] = bwboundaries(BW);
%get stats
stats= regionprops(L, 'Centroid', 'Area', 'Perimeter');
Centroid = cat(1, stats.Centroid);
Perimeter = cat(1,stats.Perimeter);
Area = cat(1,stats.Area);
CircleMetric = (Perimeter.^2)./(4*pi*Area); %circularity metric
SquareMetric = NaN(N,1);
TriangleMetric = NaN(N,1);
%for each boundary, fit to bounding box, and calculate some parameters
for k=1:N,
boundary = B{k};
[rx,ry,boxArea] = minboundrect( boundary(:,2), boundary(:,1)); %x and y are flipped in images
%get width and height of bounding box
width = sqrt( sum( (rx(2)-rx(1)).^2 + (ry(2)-ry(1)).^2));
height = sqrt( sum( (rx(2)-rx(3)).^2+ (ry(2)-ry(3)).^2));
aspectRatio = width/height;
if aspectRatio > 1,
aspectRatio = height/width; %make aspect ratio less than unity
end
SquareMetric(k) = aspectRatio; %aspect ratio of box sides
TriangleMetric(k) = Area(k)/boxArea; %filled area vs box area
end
%define some thresholds for each metric
%do in order of circle, triangle, square, rectangle to avoid assigning the
%same shape to multiple objects
isCircle = (CircleMetric < 1.1);
isTriangle = ~isCircle & (TriangleMetric < 0.6);
isSquare = ~isCircle & ~isTriangle & (SquareMetric > 0.9);
isRectangle= ~isCircle & ~isTriangle & ~isSquare; %rectangle isn't any of these
%assign shape to each object
whichShape = cell(N,1);
whichShape(isCircle) = {'Circle'};
whichShape(isTriangle) = {'Triangle'};
whichShape(isSquare) = {'Square'};
whichShape(isRectangle)= {'Rectangle'};
%now label with results
RGB = label2rgb(L);
imshow(RGB); hold on;
Combined = [CircleMetric, SquareMetric, TriangleMetric];
for k=1:N,
%display metric values and which shape next to object
Txt = sprintf('C=%0.3f S=%0.3f T=%0.3f', Combined(k,:));
text( Centroid(k,1)-20, Centroid(k,2), Txt);
text( Centroid(k,1)-20, Centroid(k,2)+20, whichShape{k});
end

  10 Comments

Show 7 older comments
Priya Raj
Priya Raj on 20 Mar 2018
Sir,the above code is working for the image given in the code but it is not working for the other images, the value of N is not getting correctly and the combined value obtained by the metrics is also not getting in a correct form. ty

Sign in to comment.

More Answers (10)

Image Analyst
Image Analyst on 19 Feb 2014
Edited: Image Analyst on 19 Mar 2016
Look at the perimeter squared to area ratio. Use regionprops as shown in my Image Segmentation Tutorial: http://www.mathworks.com/matlabcentral/fileexchange/?term=authorid%3A31862
[EDIT]
See attached demo.

  19 Comments

Image Analyst
Image Analyst on 17 Jan 2020
You forgot to attach pic.jpeg, and your m-file. I'll look at them after you attach them.
Image Analyst
Image Analyst on 18 Jan 2020
There is NO reason for you to use edge detection on those images. I don't know why all beginners think edge detection is the first step in any image processing problem just because an image has sharp edges in it. It is almost always NOT the best first step. What you need to do is first get a binary image of the shapes and for those computer graphic images you simply want to threshold.
For the monochrome image:
binaryImage = grayScaleImage < 128;
For the color image
% Find out where any pixel is not white.
binaryImage = min(rgbImage, [], 3) < 255;

Sign in to comment.


phaneendra ch
phaneendra ch on 22 Nov 2015
While executing the above code I am getting an error in the code I.e.,(undefined function or variable 'minbounderect'). As I am new to MATLAB plz solve this prblm.tq in advance sir

  2 Comments

kaz
kaz on 25 Apr 2016
i am so noob with matlab. sir how to add this function to the codes at the top. which i am getting the same error minboundrect. ty

Sign in to comment.


Venkatesh A
Venkatesh A on 12 Dec 2015
Mr.matt kindig, I am working with your "various shapes detecting" code which is in this page. The code u have written is working very well for the above black and white image(the image which you chosen to write code). But I am using some other images( 'shapes.jpg' which is now I am submitting) which might have some holes in the inside of the image and I am getting error( error is the matrix cannot be generated). How to fill the small holes in that image. Can I draw subplots for your code for various shapes that are present in the image?

  1 Comment

Image Analyst
Image Analyst on 12 Dec 2015
Venkatesh A, you forgot to give the link to your question where you attached your code and image.
If you do that, we can go to your question and tell you how to use imfill(binaryImage, 'holes') to fill holes in your image.

Sign in to comment.



AKHIL RAJAGOPAL
AKHIL RAJAGOPAL on 23 Apr 2016
I am getting an error at isTriangle = ~isCircle & (TriangleMetric < 0.6); as: Error using & Matrix dimensions must agree.
Please help me solve this problem.

  10 Comments

Show 7 older comments
kaz
kaz on 25 Apr 2016
i do wonder if you have another demo that can measure rectangular shape in cm. of course with know object dimension in the real life. for example in the image attached. the red rectangular is 3cm X 3cm. how would i be able to find the frame size with knowing the dimensions of the box inside it.

Sign in to comment.


noor jahan m
noor jahan m on 8 Dec 2016
i am a beginner in matlab. I tried the code for rectangle detection. I cud also find the function minboundrect . but i got error in convex hull of this function. Can u please help further? thank u

  0 Comments

Sign in to comment.


Rahul Chauhan
Rahul Chauhan on 23 Oct 2017
Ty very much sir for the code but I'm getting error as "undefined function or variable 'minboundract'" So plz help me sir getting this error correct asap.... Again Ty in advance sir

  4 Comments

Show 1 older comment
Rahul Chauhan
Rahul Chauhan on 24 Oct 2017
Ty sir but again showing error as convhull and also which values are to be supplied at minboundract.m and when I running it it shows provide values as in run button in maltab like-minded Run :minboundract (x,y, metric) so know which values I should provide here...... Help me sir about this problem...
Image Analyst
Image Analyst on 24 Oct 2017
Again, it's minboundrect(), not minboundract().
Supply your image and code in a new question (not as an Answer here in ROMIL's discussion thread - he probably doesn't care anymore since he posted this three and a half years ago.)
Rahul Chauhan
Rahul Chauhan on 25 Oct 2017
again sir i corrected the function but the code is not working here i am attaching the code and image file know help me.... here is the code :(1)file attached as test.m
(2)minboundrect.m
here is the image: abc.jpg
ty in advance sir for helping me.....

Sign in to comment.


Pavel Vilbik
Pavel Vilbik on 11 Dec 2017
How to find the qr code( this plastic card) in this photo, as it has a begining job, I need a coordinate and axis, and that I would be marked with a ractangle.

  1 Comment

Image Analyst
Image Analyst on 11 Dec 2017
You should start your own question on this. In that question I'll tell you how to find the blobs, perhaps based on color saturation, and then to take the histogram of each blob looking for a fairly bimodal histogram. The standard deviation of the histogram of the all black and all white objects will be much less than a checkerboard object. If you still can't figure it out, I might post code over in that new question you're going to post.

Sign in to comment.


Michelle de Bock
Michelle de Bock on 28 Dec 2018
Hi Sir,
Is it also possible to classify the direction of the triangle. e.g. left pointing triangle or right pointing triangle? Also classifying the thrid/fourth object in the picture? This is not really a rectangle, but how to separate this from a real rectangle?
Kind regards,
Michelle

  1 Comment

Image Analyst
Image Analyst on 28 Dec 2018
Yes, I'm sure you could. Just modify my attached shape recognition demo. Once you have the blob, find its bounding box and centroid with regionprops. Then if the centroid is to the left of the centerline of the bounding box, it's pointing to the left. If it's below, it's pointing up.
For the other object, you'll also have to look for how many vertices it has and then perhaps scale a template to its size and see if enough pixels match to be considered that object. You could also do the template matching method with the triangles if you want. No, I don't have code for that but, being smart engineer, I'm sure you will find it easy to do.

Sign in to comment.


Ahaana Khurana
Ahaana Khurana on 4 Feb 2020
ROMIL can you pls provide the code for SHAPE DETECTION on ahaanakhurana@gmail.com.

  3 Comments

Dina Abd El-twab
Dina Abd El-twab on 24 Feb 2020
pp=alexnet;
ppl=pp.Layers;
pp=pp.Layers(1:19);
ppp=[pp
fullyConnectedLayer(2)
softmaxLayer()
classificationLayer()]
options = trainingOptions('sgdm',...
'InitialLearnRate',1e-3,...
'MaxEpochs',10,...
'CheckpointPath',tempdir);
train1 = trainFasterRCNNObjectDetector(gTruth,ppp,options, ...
'NegativeOverlapRange',[0 0.1], ...
'PositiveOverlapRange',[0.5 1], ...
'SmallestImageDimension',300);
a = imread('US0018_0131.png');
a = imresize(a,[227 227]);
[bbox,score,label] = detect(train1,a);
detect= insertShape(a,'rectangle',bbox);
figure
imshow(detect)
@Image Analyst
Image Analyst
Dina Abd El-twab
Dina Abd El-twab on 24 Feb 2020
I applied this code to draw a rectangle on the region of interest that i want after taining using faster RCNN .I want to convert the drawn rectangle to be circle in the next step , could you help me please ?
@Image Analyst
Image Analyst

Sign in to comment.