Skip to content

aditipanda/dconv-elunn

Repository files navigation

dconv-elunn

This is the official implementation of the gray image denoising model proposed in the paper "Exponential linear unit dilated residual network for digital image denoising".

If you use any part of this code, please cite the paper using this:

@article{10.1117/1.JEI.27.5.053024, author = {Aditi Panda and Ruchira Naskar and Snehanshu Pal}, title = {{Exponential linear unit dilated residual network for digital image denoising}}, volume = {27}, journal = {Journal of Electronic Imaging}, number = {5}, publisher = {SPIE}, pages = {1 -- 14}, keywords = {convolutional neural networks, deep learning, image denoising, image enhancement, Image denoising, Denoising, Convolution, Network >architectures, Data modeling, Image filtering, Image processing, Image quality, Performance modeling, Visualization}, year = {2018}, doi = {10.1117/1.JEI.27.5.053024}, URL = {https://www.spiedigitallibrary.org/journals/journal-of-electronic-imaging/volume-27/issue-5/053024/Exponential-linear-unit-dilated-residual-network-for-digital-image-denoising/10.1117/1.JEI.27.5.053024.short?SSO=1} }

The original paper contains two different variants for gray image denoising - a 10-layer network and a 5-layer network. Here, the 5-layer variant is presented.