Improvement of Luminance Isotropy for Convolutional Neural Networks-Based Image Super-Resolution

被引：1

作者：

Urazoe, Kazuya ^{[1
]}

Kuroki, Nobutaka ^{[1
]}

Kato, Yu ^{[1
,2
]}

Ohtani, Shinya ^{[1
,3
]}

Hirose, Tetsuya ^{[1
,4
]}

Numa, Masahiro ^{[1
]}

机构：

[1] Kobe Univ, Grad Sch Engn, Kobe, Hyogo 6578501, Japan

[2] EIZO Corp, Haku San, Japan

[3] Toyota Motor Co Ltd, Toyota, Japan

[4] Osaka Univ, Grad Sch Engn, Suita, Osaka, Japan

来源：

IEICE TRANSACTIONS ON FUNDAMENTALS OF ELECTRONICS COMMUNICATIONS AND COMPUTER SCIENCES | 2020年 / E103A卷 / 07期

关键词：

super-resolution; resolution enhancement; convolutional neural network; isotropy; deep learning;

D O I：

10.1587/transfun.2019EAL2168

中图分类号：

TP3 [计算技术、计算机技术];

学科分类号：

0812 ;

摘要：

Convolutional neural network (CNN)-based image superresolutions are widely used as a high-quality image-enhancement technique. However, in general, they show little to no luminance isotropy. Thus, we propose two methods, "Luminance Inversion Training (LIT)" and "Luminance Inversion Averaging (LIA)," to improve the luminance isotropy of CNN-based image super-resolutions. Experimental results of 2 x image magnification show that the average peak signal-to-noise ratio (PSNR) using Luminance Inversion Averaging is about 0.15-0.20 dB higher than that for the conventional super-resolution.

引用

页码：955 / 958

页数：4

共 50 条

[21] Image Super-Resolution Based on Error Compensation with Convolutional Neural Network
Lu, Wei-Ting
Lin, Chien-Wei
Kuo, Chih-Hung
Tung, Ying-Chan
[J]. 2017 ASIA-PACIFIC SIGNAL AND INFORMATION PROCESSING ASSOCIATION ANNUAL SUMMIT AND CONFERENCE (APSIPA ASC 2017), 2017, : 1160 - 1163
[22] Image Super-resolution Reconstruction Algorithm Based on Convolutional Neural Network
He Jingxuan
Zhang Jian
Zhang Yonghui
Wang Rong
[J]. PROCEEDINGS OF 2018 IEEE INTERNATIONAL CONFERENCE ON AUTOMATION, ELECTRONICS AND ELECTRICAL ENGINEERING (AUTEEE), 2018, : 267 - 271
[23] Single Image Super-Resolution Using Frequency - Dependent Convolutional Neural Networks
Baek, Sangwook
Lee, Chulhee
[J]. 2020 IEEE INTERNATIONAL CONFERENCE ON INDUSTRIAL TECHNOLOGY (ICIT), 2020, : 692 - 695
[24] License Plate Image Super-Resolution Based on Convolutional Neural Network
Yang, Yang
Bi, Ping
Liu, Ying
[J]. 2018 IEEE 3RD INTERNATIONAL CONFERENCE ON IMAGE, VISION AND COMPUTING (ICIVC), 2018, : 723 - 727
[25] Terahertz image super-resolution based on a complex convolutional neural network
Wang, Ying
Qi, Feng
Wang, Jinkuan
[J]. OPTICS LETTERS, 2021, 46 (13) : 3123 - 3126
[26] TISR: Twin Image Super-Resolution using Deep Convolutional Neural Networks
Muhammad, Wazir
Bhutto, Zuhaibuddin
Shah, Jalal
Shaikh, Murtaza Hussain
Shah, Syed Ali Raza
Butt, Shah Muhammad
Masroor, Salman
Hussain, Ayaz
[J]. INTERNATIONAL JOURNAL OF COMPUTER SCIENCE AND NETWORK SECURITY, 2022, 22 (01): : 443 - 448
[27] Image Fusion and Super-Resolution with Convolutional Neural Network
Zhong, Jinying
Yang, Bin
Li, Yuehua
Zhong, Fei
Chen, Zhongze
[J]. PATTERN RECOGNITION (CCPR 2016), PT II, 2016, 663 : 78 - 88
[28] Terahertz image super-resolution based on a deep convolutional neural network
Long, Zhenyu
Wang, Tianyi
You, Chengwu
Yang, Zhengang
Wang, Kejia
Liu, Jinsong
[J]. APPLIED OPTICS, 2019, 58 (10) : 2731 - 2735
[29] Improved image super-resolution algorithm based on convolutional neural network
[J]. 1600, Chinese Optical Society (37):
[30] Image Super-resolution Based on Tiny Recurrent Convolutional Neural Network
Ma Hao-yu
Xu Zhi-hai
Feng Hua-jun
Li Qi
Chen Yue-ting
[J]. ACTA PHOTONICA SINICA, 2018, 47 (04)

← 1 2 3 4 5 →