Algorithm advice: defect detection

asked 2017-04-23 14:37:09 -0500

taylorty gravatar image

updated 2017-04-24 08:34:42 -0500

I want to detect defects on the surface of a cylinder object. This image shows the object without defects. This image shows the object with defects. What would be an optimal algorithm to choose in this case? Should I use deep learning instead of OpenCV?

Any advice is appreciated.

edit retag flag offensive close merge delete



For such a constrained application, deep learning is overkill. Is the cylinder always positioned in the exact same way? And is the image always from the exact same location?

StevenPuttemans gravatar imageStevenPuttemans ( 2017-04-24 08:36:20 -0500 )edit

The cylinder is not positioned in the exact same way and there are many square holes on the surface as you can see. They are not going to be the same the whole time. I tried to subtract the difference of these two images and they turned out to be very different.

taylorty gravatar imagetaylorty ( 2017-04-27 02:46:11 -0500 )edit