Image super-resolution reconstruction based on improved generative adversarial network

被引：0

作者：

Wang Y.-L. ^{[1
]}

Li X.-J. ^{[1
]}

Ma H.-B. ^{[1
]}

Ding Q. ^{[1
]}

Pirouz M. ^{[3
]}

Ma Q.-T. ^{[2
]}

机构：

[1] Department of Electronic Engineering, Heilongjiang University, No.74 Xuefu Road, Harbin, Heilongjiang

[2] Faculty of Engineering, The Hong Kong Polytechnic University, 11 Yucai Road, Hung Hom, Kowloon

[3] Department of Computer Science, California State University, Fresno

来源：

Journal of Network Intelligence | 2021年 / 6卷 / 02期

关键词：

Convolutional neural network; Generative adversarial network; Image super-resolution; Recursive residual network;

D O I：

10.1109/ICIAI.2019.8850808

中图分类号：

学科分类号：

摘要：

This article introduces the super-resolution reconstruction of images based on an improved generative confrontation network, improves the network structure of the generator, and proposes a super-resolution reconstruction algorithm for the recursive residual generation confrontation network. Its discriminator uses PatchGAN as the discriminator. The network solves the bottleneck of low feature information utilization and slow convergence of the generation countermeasure network in the super-resolution algorithm based on convolutional neural network. The reconstruction algorithm is compared with mainstream super-resolution reconstruction algorithms on standard data sets such as Set5 and Set14. The data shows that the algorithm can effectively improve the use of feature information, restore the details of low-resolution images, and improve the quality of image reconstruction. © 2021, Taiwan Ubiquitous Information. All rights reserved.

引用

页码：155 / 163

页数：8

共 19 条

[1]

Xing Z. Y., Xiao E. L., Jian X. Z., Infrared image super-resolution reconstruction of dual discriminant generative confrontation network[J], Small Microcomputer System, 41, 5, pp. 662-667, (2020)

[2]

Wang Y. N., Li W. T., Ren J., Single-frame image super-resolution algorithm based on generative confrontation network [J], Foreign Electronic Measurement Technology, 39, 1, pp. 26-32, (2020)

[3]

Li C., Zhang Y., Huang C. H., Improved generation of confrontation network image super-resolution reconstruction[J], Computer Engineering and Applications, 56, 4, pp. 191-196, (2020)

[4]

Dong C., Loy C. C., He K., Et al., Image super-resolution using deep convolutional networks[J], IEEE Transactions on Pattern Analysis and Machine Intelligence, 38, 2, pp. 295-307, (2015)

[5]

Kim J., Lee J. K., Lee K. M., Accurate image super-resolution using very deep convolutional networks[C], IEEE Conference on Computer Vision and Pattern Recognition, pp. 1646-1654, (2016)

[6]

Ledig C., Theis L., Huszar F., Et al., Photo-realistic single image super-resolution using a generative adversarial network[C], IEEE Conference on Computer Vision and Pattern Recognition, pp. 4681-4690, (2017)

[7]

Lim B., Son S., Kim H., Et al., Enhanced deep residual networks for single image super-resolution[C], IEEE Conference on Computer Vision and Pattern Recognition Workshops, pp. 136-144, (2017)

[8]

Tseng K. K., Zhang R., Chen C. M., Hassan M. M., DNetUnet: a semi-supervised CNN of medical image segmentation for super-computing AI service, The Journal of Supercomputing, (2020)

[9]

Wang E. K., Chen C. M., Hassan M. M., Almogren A., A deep learning based medical image segmentation technique in Internet-of-Medical-Things domain, Future Generation Computer Systems, 108, pp. 135-144, (2020)

[10]

Wang E. K., Chen C. M., Wang F., Khan M. K., Kumari S., Joint-learning segmentation in Internet of drones (IoD)-based monitor systems, Computer Communications, 152, pp. 54-62, (2020)

← 1 2 →