Noise filtering with the Ungerflow algorithm [1]. More...
Detailed Description
Noise filtering with the Ungerflow algorithm [1].
Usage: ungerflow( (char_image)source and sink, (float_image)gradient image, (float_image)lambda image, (float_image)guidence image, (int)number of iterations, (float)tau, (float)theta) -> float_image
Description:
The input image is noise filtered according to the following relation:
![\[ \min\limits_{u, v}\left\{ \int\limits_\Omega g(x)|\nabla u|d\Omega + \frac{1}{\theta}\int\limits_\Omega(u-v)^2d\Omega + \int\limits_\Omega \lambda(x)|v-f|d\Omega \right\} \]](form_0.png)
Types supported: float 2d, float 3d, float 4d, ..., float xd
Category: signal, development ,
References:
[1] Horst Bischof Markus Unger, Thomas Pock. Interactive globally optimal image segmentation. Technical Report ICGTR08/02, Institute for Computer Graphics and Vision Graz University of Technology, Austria, July 2008.
[2] M. Unger, T. Pock, and H. Bischof. Continuous globally optimal image segmentation with local constraints. In Computer Vision Winter Workshop 2008, Moravske Toplice, Slovenija, February 2008.
Examples: can be found in 'tutorial/python/ungerflow'
