Hi Pilgun

Thanks for your response. I have included this image as an input image. Will this work out. Moreover, can you kindly specify what sort of outputs are you talking about. Is it the magnitude and the orientation?

Hi Pilgun I have studied this paper and I have used the roundness equation. I took the matlab reference code from here https://www.mathworks.com/matlabcentral/fileexchange/45961-fast-radial-symmetry-transform/content/frst2d.m

As you can see the input picture consists of different shape objects so I initially need to detect the central regions of the each leaves. 

Hi Reichl 

Thanks for your response. Can you share some examples so that I can get a more clear picture. Moreover are you trying to explain the skeleton as a morphological operation.

Dear Dheeman Saha ,

Look the links. May be useful.

http://lmb.informatik.uni-freiburg.de/people/bahlmann/data/ni_si_ba_cvpr12.pdf

http://www2.it.lut.fi/project/comphi1/files/thesis_zafari.pdf

Regards, Shafagat

Dear DR. Mahmudova

I looked into those references before the implementation and that was helpful. Thank you.

Dear Reichl

So the termal regions are the center of the all the regions of the plant. But Reichl here its my idea. I want to just find the termal region of the elliptic shape leaves so that I can apply the Fast Radial Symmetry can be used. More can you kindly share how you precessed the thermal regions.

Dheeman

So do you think that the local maxima should be somewhat related to the seeding point. Moreover can you kindly share the link where you used the algorithm for marking the thermal points. Will the Euclidean Distance work as a local maxima.