Deep Convolutional Neural Network Based on Multi-Scale Feature Extraction for Image Denoising

With the development of deep learning, many methods on image denoising have been proposed processing images on a fixed scale or multi-scale which is usually implemented by convolution or deconvolution. However, excessive scaling may lose image detail information, and the deeper the convolutional net...

Full description

Saved in:
Bibliographic Details
Published in2020 IEEE International Conference on Visual Communications and Image Processing (VCIP) pp. 213 - 216
Main Authors Zhang, Jing, Sang, Liu, Wan, Zekang, Wang, Yuchen, Li, Yunsong
Format Conference Proceeding
LanguageEnglish
Published IEEE 01.12.2020
Subjects
convolutional network
Diamond
Feature extraction
Image denoising
Image reconstruction
multi-scale
Noise level
Noise reduction
Training
Online AccessGet full text

Cover

More Information
Summary:With the development of deep learning, many methods on image denoising have been proposed processing images on a fixed scale or multi-scale which is usually implemented by convolution or deconvolution. However, excessive scaling may lose image detail information, and the deeper the convolutional network the easier to lose network gradient. Diamond Denoising Network (DmDN) is proposed in this paper, which mainly based on a fixed scale and meanwhile considering the multi-scale feature information by using the Diamond-Shaped (DS) module to deal with the problems above. Experimental results show that DmDN is effective in image denoising.
ISSN:2642-9357
DOI:10.1109/VCIP49819.2020.9301843