1
$\begingroup$

Are there any projects where you can detect and measure the width of a crack? I am using tensorflow and labeling the data sets for now.

$\endgroup$
  • $\begingroup$ Although I love the conceptual center of the question, which is optical gap measurement, a key technology area that certainly benefits from computer vision and AI, I'm voting to close this question as off-topic because of the convention across SE communities to discourage requests for implementation resources, essentially downloads. (See ai.meta.stackexchange.com/questions/1267/…) $\endgroup$ – Douglas Daseeco Jan 14 at 4:43
  • $\begingroup$ ai.stackexchange.com/questions/9971/… was created to hopefully provide a more sustainable location for good answers if this question is closed. $\endgroup$ – Douglas Daseeco Jan 14 at 4:44
2
$\begingroup$

Depending on the nature of your images, the usual method for doing this sort of thing would be to use an edge detection filter on your image and look for pairs of parallel lines. Such lines might imply the presence of a crack. A Hough transform on the filtered image would then provide the position and orientation of the crack. A convolutional neural network could certainly learn to detect edges (and hence cracks), which might be all that you need.

$\endgroup$

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.