0

Currently, the Scikit image package implements Laplacian of Gaussian (LoG) to find spherical blobs. The package is here. How can one convert these Laplacian of Gaussian to an elliptical Gaussian kernel? This is required to get the non-spherical blobs in an image.

Tara Mishra
  • 154
  • 8
  • The min_sigma, max_sigma of LoG can take a sequence of scalars, so you can specify the desired values for each axis of your ellipse. Check the [LoG documentation](https://scikit-image.org/docs/dev/api/skimage.feature.html#skimage.feature.blob_log). – dhanushka Dec 22 '20 at 02:27
  • This doesn't check for all possible combination of sigma. For example if I give, min_sigma = [3,4] , max_sigma = [10,11] and num_sigma = 7. The LoG implemented in scikit-image only considers sigma (x,y) values such as (3,4), (4,5), (5,6) and so on. However, I also want sigma values of (3,5), (3,6), (5,3), etc. – Tara Mishra Dec 28 '20 at 08:00

0 Answers0