Deep Against Net Image Super-Resolution Reconstruction Algorithm Based on W Distance

DOI

10.2991/ncce-18.2018.128How to use a DOI?

Keywords

super-resolution; GAN; Wasserstein distance

Abstract

This paper proposes an image super-resolution algorithm based on the condition of Wasserstein distance generation against network, aiming at improving the quality of reconstructed images and improving the condition against the stability of neural network in image super-resolution research. Wasserstein distance algorithm is used to solve the instability problem of the traditional GAN network generator and make the model more stable. In the test sets such as Set5 and Set14, the three evaluation indexes PSNR, SSIM, and IFC of the SRWCGAN algorithm are superior to the VDSR algorithm, and the convergence speed and stability are better than those without W-distance.

Copyright

© 2018, the Authors. Published by Atlantis Press.

Open Access

This is an open access article distributed under the CC BY-NC license (http://creativecommons.org/licenses/by-nc/4.0/).

Download article (PDF)

TY  - CONF
AU  - Yang Liu
PY  - 2018/05
DA  - 2018/05
TI  - Deep Against Net Image Super-Resolution Reconstruction Algorithm Based on W Distance
BT  - Proceedings of the 2018 International Conference on Network, Communication, Computer Engineering (NCCE 2018)
PB  - Atlantis Press
SP  - 780
EP  - 783
SN  - 1951-6851
UR  - https://doi.org/10.2991/ncce-18.2018.128
DO  - 10.2991/ncce-18.2018.128
ID  - Liu2018/05
ER  -