-
Notifications
You must be signed in to change notification settings - Fork 1
Gimp Plugin using CUDA: Multiresolution Gradient Adaptive Filter
License
richardmembarth/gimp-cuda-plugin
Folders and files
Name | Name | Last commit message | Last commit date | |
---|---|---|---|---|
Repository files navigation
Gimp Plugin: Multiresolution gradient adaptive filter ===================================================== The gimp plugin implements a multiresolution gradient adaptive filter, as described in [1], however the filter kernel was replaced by a bilateral filter, as described in [2]. The major advantage of this filter is the fact that the filter preserves edges while noise is removed. The multiresolution filter consists of 3 major steps (see also doc/mrfilt.eps): 1. Decompose: the gaussian (g0) and laplace (l0) pyramid are constructed 2. Filter: the bilateral filter (f0) gets applied to the laplace pyramid 3. Reconstruct: the output image (r0) is reconstructed from the filted data The multiresolution filter operates on gray-scale images and requires the image resolution to be a power of 2. In case that the image doesn't fulfill this requirement, the image is padded with mirrored pixel values. The multiresolution filter works on 5 stages and the parameters for the bilateral filter kernel can be adjusted (see also: INSTALL file). --- [1] D. Kunz, K. Eck, H. Fillbrandt, and T. Aach. Nonlinear multiresolution gradient adaptive filter for medical images. Proceedings of SPIE, 5032:732, 2003. [2] C. Tomasi and R. Manduchi. Bilateral filtering for gray and color images. Proceedings of the Sixth International Conference on Computer Vision, page 839, 1998.
About
Gimp Plugin using CUDA: Multiresolution Gradient Adaptive Filter
Resources
License
Stars
Watchers
Forks
Releases
No releases published
Packages 0
No packages published