Identifying Bulge in Image
3 views (last 30 days)
Show older comments
Chinna Babu
on 26 Mar 2021
Commented: Image Analyst
on 29 Mar 2021
Hello,
I have bunch of images, some images have filaments with a bulge (as shown in Pic1.jpg) and some images have filaments with no bulge (as shown in Pic2.jpg).I need to go through each filament in an image and identify if the filament has a bulge or not. How can I identify a bulge or stright line in the image? Can anyone kindly help?
Thanks!
1 Comment
Sulaymon Eshkabilov
on 26 Mar 2021
If the sizes of your images are the same or different. If they have the same sizes, then you can employ logical indexing by comparing several rows of the image against its threshold values to estimate whether the image has a buldge or no.
Accepted Answer
Image Analyst
on 27 Mar 2021
Attached is what I have so far but it's not perfect. You really need to make sure that all your tubes fit inside the field of view, even when bulged. If you do that, it could be made perfect. Please attach images where that is the case.
2 Comments
Image Analyst
on 29 Mar 2021
Chinna, like I said, you should fix your immage capture conditions. It will be much easier to do that than to try tomodify the algorithm to handle umpteen bazillion ways that your image can be bad or corrupted. Choose how many filaments will be in your field of view and stick with it. And make sure the field of view is wide enough to contain any bulges. Then we will not have to spend hours trying to develope an algorithm to handle bad images.
More Answers (1)
Image Analyst
on 26 Mar 2021
Easy. Threshold. Then take the two largest blobs. Then use regionprops to determine their area. If it's less than the "no bulge" area, then there's a bulge protruding into them.
0 Comments
See Also
Categories
Find more on Deep Learning Toolbox in Help Center and File Exchange
Products
Community Treasure Hunt
Find the treasures in MATLAB Central and discover how the community can help you!
Start Hunting!