+- Python Forum (https://python-forum.io)
+-- Forum: Python Coding (https://python-forum.io/forum-7.html)
+--- Forum: General Coding Help (https://python-forum.io/forum-8.html)
+--- Thread: Denoise image (/thread-25186.html)
Denoise image - smitha - Mar-23-2020
I want to implement non-local total variation for gamma noise removal using a gradient descent approach( for denoising a noisy image). Can anyone please guide me on the relevant functions available either using open cv, skimage, PyTorch or using TensorFlow?
I am aware of the built-in functions "skimage.restoration.denoise_tv_chambolle" and "skimage.restoration.denoise_tv_bregman". However, they are not non-local methods. Since I want to modify these codes (it is not possible to modify original codes, since it has .so files), I have to implement the same functionality for non-local image processing.
Thank you in advance for your help
RE: Denoise image - ndc85430 - Mar-23-2020
(Mar-23-2020, 05:42 AM)smitha Wrote: (it is not possible to modify original codes, since it has .so files)
I'm not sure why that's a problem really. scikit-learn is open source, so all the source code is available (https://github.com/scikit-learn/scikit-learn). I'd encourage you to fork it, make the changes you want and then contribute them back to the project.
RE: Denoise image - smitha - Mar-23-2020
Thank you for sending the github editable link. However, the function that I am looking for is https://github.com/scikit-image/scikit-image/blob/master/skimage/restoration/non_local_means.py#L11
Here I could not find the source code of the line " from ._nl_means_denoising import ". Kindly request you to help me get the source code of the same.