7

This is the thresholded image in context. The part that is unwanted is the extension on the top, I hope it's deducible. If it isn't, please ask. I want the almost circular part that would remain after removing the extension in context. How do I go about doing it? I've had someone suggest morphologial operation would help. Could someone please expand on that, and maybe provide an example?

Thanks

enter image description here

Here's another example:

enter image description here

BoltClock
  • 700,868
  • 160
  • 1,392
  • 1,356

3 Answers3

5

The best practice would be, since you only need shapes similar to circle. Try using curve fitting along with Kalman Filter. Hope this would definitely tackle the problem in a more efficient way. circle equation will try to pull(correct) the points towards the center and Kalman will help trace the curve effectively.

Anubhav Rohatgi
  • 458
  • 1
  • 6
  • 16
0

Your image is like a piece of flat plane3D view of your image . OTSU is not always a good idea. Disable it and choose a manual threshold (75).

LBerger
  • 593
  • 2
  • 12
0
from scipy import misc
from skimage.morphology import erosion    
data = misc.imread('gycl1.jpg')
eroded = erosion(data, disk(25)

enter image description here

Is this what you meant? I am using the erosion morphological operator with a disk size (to maintain the circular function).

rmkemker
  • 470
  • 5
  • 16