Hi Vighnesh, I'm not familiar with blob detection algorithms, but I will say that in general in scikit-image we try to shy away from new classes, and prefer to represent things with numpy arrays wherever possible. I think an integer image with one labeled region per blob would be appropriate (and could be piped into regionprops). Juan. On Thu, Feb 27, 2014 at 5:41 PM, Vighnesh Birodkar < vighneshbirodkar@gmail.com> wrote:
Hello
I am applying as a GSoC student for scikit-image this year. Last year I was a GSoC student for SimpleCV. To start contributing I wanted to implement Blob Detection ( DOG and LOG initially ).
I wanted to have some guidelines as to where to put the code and how to structure it.Is there any piece of code lying around in scikit-image which I can use as a template for Blob Detection ?
I would put the new code inside a new .py file inside features. I think it would be a good idea to keep the prototypes identical for all blob detection algorithms. My first intuition is to return a class ( or a dictionary ) with some attributes set. Some blob detection algorithms may compute the area of the blobs as well while some might not. Or I can just return the (x,y) coordinates.
Please point out if I am mistaken somewhere.
Thanks Vighnesh
-- You received this message because you are subscribed to the Google Groups "scikit-image" group. To unsubscribe from this group and stop receiving emails from it, send an email to scikit-image+unsubscribe@googlegroups.com. For more options, visit https://groups.google.com/groups/opt_out.