Originally cloned from this repo. Made stable for python3 without errors. Works with tf >= 2.
A tensorflow implement of the TIP2017 paper Beyond a Gaussian Denoiser: Residual Learning of Deep CNN for Image Denoising
- BSD68 Average Result
The average PSNR(dB) results of different methods on the BSD68 dataset.
| Noise Level | BM3D | WNNM | EPLL | MLP | CSF | TNRD | DnCNN-S | DnCNN-B | DnCNN-tensorflow |
|---|---|---|---|---|---|---|---|---|---|
| 25 | 28.57 | 28.83 | 28.68 | 28.96 | 28.74 | 28.92 | 29.23 | 29.16 | 29.17 |
- Set12 Average Result
| Noise Level | DnCNN-S | DnCNN-tensorflow |
|---|---|---|
| 25 | 30.44 | 30.38 |
tensorflow >= 2
numpy
opencv
I used the BDS500 dataset for training, you can download it here: http://www.eecs.berkeley.edu/Research/Projects/CS/vision/grouping/BSR/BSR_bsds500.tgz It contains 500 RGB images, 400 for training and 100 for testing.
Before training, you have to rescale the images to 180x180 and adding noise to them. The folder structure is supposed to be:
./data/train/original for the 180x180 original train images
./data/train/noisy for the 180x180 noisy train images
./data/test/original for the 180x180 original test images
./data/test/noisy for the 180x180 noisy test images
You need the original files for testing just to calculate the PSNR. You can denoise without original files: just put the noisy files also in ./data/test/original .
$ python main.py
(note: You can add command line arguments according to the source code, for example
$ python main.py --batch_size 64 )
if you wish to check out if it is working, copy the contents of checkpoint_demo to checkpoint folder and execute:
./fast_starter.sh
$ python main.py --phase test