The Restoration of Style Chinese Characters Based on Deep Learning

DOI

10.2991/ncce-18.2018.67How to use a DOI?

Keywords

Deep learning; generative adversarial network; structure generated clear.

Abstract

For the image inpainting using incomplete Chinese characters proposed a new method for Chinese characters repair, the first use of U-Net type network structure combining the training method of generating against network design a style Chinese characters converter, and then repair the missing content by Chinese characters style converter. The experimental results show that the structure of the Chinese characters for the repair of the clear, less noise, edge connection also smooth, look real. This paper provides a new idea for image inpainting, it can be converted from other images to the same style of image for image recovery.

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  - Da Lv
AU  - Yijun Liu
PY  - 2018/05
DA  - 2018/05
TI  - The Restoration of Style Chinese Characters Based on Deep Learning
BT  - Proceedings of the 2018 International Conference on Network, Communication, Computer Engineering (NCCE 2018)
PB  - Atlantis Press
SP  - 426
EP  - 430
SN  - 1951-6851
UR  - https://doi.org/10.2991/ncce-18.2018.67
DO  - 10.2991/ncce-18.2018.67
ID  - Lv2018/05
ER  -