OS9-7 Research on Image Super-Resolution Reconstruction Based on Deep Learning

Title:	OS9-7 Research on Image Super-Resolution Reconstruction Based on Deep Learning
Publication:	ICAROB2020
Volume:	25
Pages:	640-643
ISSN:	2188-7829
DOI:	10.5954/ICAROB.2020.OS9-7
Author(s):	Lingran An, Fengzhi Dai, Yasheng Yuan
Publication Date:	January 13, 2020
Keywords:	Super-resolution, deep learning, neural network, Generative Adversarial Networks
Abstract:	This paper mainly applies the relevant theories of deep learning to image super-resolution reconstruction technology. By comparing four classical network models used for image super-resolution (SR), finally a generative adversarial network (GAN) is selected to implement image super-resolution, which is called SRGAN. SRGAN consists of a generator and a discriminator that uses both perceived loss and counter loss to enhance the realism of the output image in detail. The data sets used by the training network are partly from the network and partly from the artificial. Compared with other network models, the final trained SRGAN network is above average in PSNR and SSIM values. Although it is not optimal, the output high-resolution images are the best in the subjective feelings of human eyes, and the reconstruction effect in the image details is far higher than that of other networks.
PDF File:	https://alife-robotics.co.jp/members2020/icarob/data/html/data/OS/OS9/OS9-7.pdf
Copyright:	© The authors. This article is distributed under the terms of the Creative Commons Attribution License 4.0, which permits non-commercial use, distribution and reproduction in any medium, provided the original work is properly cited. See for details: https://creativecommons.org/licenses/by-nc/4.0/

HOME