How to Create Defect Class for Image Processing?

Good day everyone, I have a code to inspect part's surface using image processing. And i want one of my output to show the scratches class such as:

class 1 scratch;

class 2 scratch;

class 3 scratch.

I've already calculate the total area of the image for this system. How can I generate the coding for this kind of output? Thank you so much for your replies.

Answer 1

[Image Analyst]

Maybe try thresholding with the 'adaptive' option in imbinarize().

Or try a bottomhat filter with imbothat().

Or try getting a standard deviation image with stdfilt() and then thresholding.

Once you have the binary image you can measure all kinds of features with regionprops, like area, perimeter, maximum length, etc. Then you can train it with the Classification Learner app to take all those features and make a prediction as to what kind of defect (Class 1 scratch, class 2 scratch, etc.) it is.

[Afrena Azman]

Thank you for your reply. Just for a knowledge, can I use something like

for X = total

if any(0<X)&&(X<100)

disp("class 1 scratch")

elseif any(101<X)&&(X<300)

disp("class 2 scratch")

elseif any(301<X)&&(X<500)

disp("class 3 scratch")

elseif any(501<X)&&(X<700)

disp("class 4 scratch")

elseif any(701<X)&&(X<900)

disp("class 5 scratch")

end

end

for this? But seems like my coding would become longer am I right

[Image Analyst]

I doubt that will work. Your scratches might have lots of gray levels in them so it might just pop into the first case. Plus, it doesn't take into account the size of the scratch, it only cares about what the gray levels are.

[Afrena Azman]

Understood sir, thank you so much for your help.

Answer 2

[yanqi liu]

yes，sir，may be use regionprops to get some properties，such as

Boundingbox，Area，Circularity

then make judge rule to classify

[Afrena Azman]

Noted that. I tried using the Classification Learner Apps and it works. Thank you so much