rectangularity to detect objects

i need to detect objects from an image using rectangularity and I do not know how to implement it please help me

Answer 1

[Matt J]

Hard to say without any example images, but pgonCorners might help:

https://www.mathworks.com/matlabcentral/fileexchange/74181-find-vertices-in-image-of-convex-polygon?s_tid=srchtitle

You can use it to find the code's idea of the 4 vertices of the objects. If the vertices form a rectangle to within your tolerances, and if there is strong overlap between that rectangle and the actual object, then you would accept the object as rectangular.

[esraa mohammed]

i need to use rectangular to detect qr code object

[Matt J]

My suggestion seems applicable to this image. The detected vertex pattern for the QR code seems very distinctive.

load Image

BW=bwpropfilt(Image,'ConvexArea',[10000,Inf]);


imshow(BW); hold on

reg=regionprops(BW,'ConvexImage','BoundingBox');

N=numel(reg);

for i=1:N
 
    img=reg(i).ConvexImage;

    V = fliplr(pgonCorners(img,4))+reg(i).BoundingBox(1:2); %vertices
    
    scatter(V(:,1),V(:,2),'filled','SizeData',80,'MarkerEdgeColor','y');
end

hold off

[esraa mohammed]

I'm sorry that I tired you. But an ERROR appears in this part

[Matt J]

It's because you didn't download pgonCorners.

[Matt J]

@Khizar Khalid What about the example code above? Or at the pgonCorners File Exchange site under the Examples tab?

Answer 2

[Image Analyst]

Lots of QR code finding routines in the File Exchange.

https://www.mathworks.com/matlabcentral/fileexchange?q=qr+code

Some of them might be fairly robust, so check them out.