How can I identify broken pieces of rice using my webcam?

I have been using this code to use my webcam to track objects in real time(the code tracks red objects only):

% Capture the video frames using the videoinput function % You have to replace the resolution & your installed adaptor name. vid = videoinput('winvideo', 1,'YUY2_352x288');

% Set the properties of the video object set(vid, 'FramesPerTrigger', Inf); set(vid, 'ReturnedColorspace', 'rgb') vid.FrameGrabInterval = 5;

%start the video aquisition here start(vid)

% Set a loop that stop after 100 frames of aquisition while(vid.FramesAcquired<=200)

    % Get the snapshot of the current frame
    data = getsnapshot(vid);

    % Now to track red objects in real time
    % we have to subtract the red component 
    % from the grayscale image to extract the red components in the image.
    diff_im = imsubtract(data(:,:,1), rgb2gray(data));
    %Use a median filter to filter out noise
    diff_im = medfilt2(diff_im, [3 3]);
    % Convert the resulting grayscale image into a binary image.
    diff_im = im2bw(diff_im,0.18);

    % Remove all those pixels less than 300px
    diff_im = bwareaopen(diff_im,60);

    % Label all the connected components in the image.
    bw = bwlabel(diff_im, 8);

    % Here we do the image blob analysis.
    % We get a set of properties for each labeled region.
    stats = regionprops(bw, 'BoundingBox', 'Centroid');

    % Display the image
    imshow(diff_im)

    hold on

    %This is a loop to bound the red objects in a rectangular box.
    for object = 1:length(stats)
        bb = stats(object).BoundingBox;
        bc = stats(object).Centroid;
        rectangle('Position',bb,'EdgeColor','r','LineWidth',2)
        plot(bc(1),bc(2), '-m+')
        %a=text(bc(1)+15,bc(2), strcat('X: ', num2str(round(bc(1))), '    Y: ', num2str(round(bc(2)))));
        %set(a, 'FontName', 'Arial', 'FontWeight', 'bold', 'FontSize', 12, 'Color', 'yellow');
    end

    hold off
end
% Both the loops end here.

% Stop the video aquisition. stop(vid);

% Flush all the image data stored in the memory buffer. flushdata(vid);

% Clear all variables clear all

I've got it from a blog, but I need to be able to identify the broken rices instead the red objects. Tried to change some lines (imsubtract) but can't make it to be what I need.

[Florian Morsch]

What do you mean by "broken rices"? Do you have rice and want to identify broken ones? How do you define broken?

[Maria Zegarra]

Hi, Yes, I have rice and want to count the smaller pieces (half, quarter of grain) I'm using a black background so it's easier to identify and quantity. But with webcam is more. Difficult

Answer 1

[Florian Morsch]

Ok, so this might help you: https://de.mathworks.com/help/vision/ug/detect-edges-in-images.html

You can try to find all grains first, then check for the pixels in each found object. If you have a set camera distance you will know how many pixel (give or take a few) one grain will have. So now you count the pixels of every found object and if its way lower then your average then it might be a broken one. If you have a moving camera you need a reference object.

This will work best if the grains are seperated and not touching each other. Otherwise two grains might be counted as one.

In this link you will find a step-by-step list what you could do: https://www.slideshare.net/SriramEmarose/bsaic-machine-vision-applications-using-mat-lab

This https://blogs.mathworks.com/steve/2008/08/20/image-visualization-using-transparency/

and this http://radio.feld.cvut.cz/matlab/toolbox/images/getting3.html might also help

[Maria Zegarra]

I will try that edge detection, so far been working with thresholding and area size. Thanks