Frequency Domain-Based Super Resolution Using Two-Dimensional Structure Consistency for Ultra-High-Resolution Display

Recent advancements in the field of super resolution (SR) have seen the adoption of generative adversarial networks (GANs) for realistic images. In this case, when performing with low-resolution (LR) images, several challenges arise due to the loss of high-frequency details from high-resolution (HR)...

Full description

Saved in:
Bibliographic Details
Main Authors: Yu Lim Seo, Suk-Ju Kang, Yeon-Kug Moon
Format: Article
Language:English
Published: MDPI AG 2024-10-01
Series:Journal of Imaging
Subjects:
Online Access:https://www.mdpi.com/2313-433X/10/11/266
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1850226883189276672
author Yu Lim Seo
Suk-Ju Kang
Yeon-Kug Moon
author_facet Yu Lim Seo
Suk-Ju Kang
Yeon-Kug Moon
author_sort Yu Lim Seo
collection DOAJ
description Recent advancements in the field of super resolution (SR) have seen the adoption of generative adversarial networks (GANs) for realistic images. In this case, when performing with low-resolution (LR) images, several challenges arise due to the loss of high-frequency details from high-resolution (HR) images, potentially leading to unwanted distortions in the generated SR images. Our paper presents a novel solution by using two-dimensional structure consistency (TSC) for image analysis. The TSC serves as a mask, enabling adaptive analysis based on the unique frequency characteristics of different image regions. Furthermore, a mutual loss mechanism, which dynamically adjusts the training process based on the results filtered by the TSC-based mask, is introduced. Additionally, the TSC loss is proposed to enhance our model capacity to generate precise TSC in high-frequency regions. As a result, our method effectively reduces distortions in high-frequency areas while preserving clarity in regions containing low-frequency components. Our method outperforms other SR techniques, demonstrating superior results in both qualitative and quantitative evaluations. Quantitative measurements, including PSNR, SSIM, and the perceptual metric LPIPS, show comparable PSNR and SSIM values, while the perceptual SR quality is notably improved according to the LPIPS metric.
format Article
id doaj-art-5cfdb1f5974d4639a93d657fbbf0ed94
institution OA Journals
issn 2313-433X
language English
publishDate 2024-10-01
publisher MDPI AG
record_format Article
series Journal of Imaging
spelling doaj-art-5cfdb1f5974d4639a93d657fbbf0ed942025-08-20T02:04:58ZengMDPI AGJournal of Imaging2313-433X2024-10-01101126610.3390/jimaging10110266Frequency Domain-Based Super Resolution Using Two-Dimensional Structure Consistency for Ultra-High-Resolution DisplayYu Lim Seo0Suk-Ju Kang1Yeon-Kug Moon2Samsung Electronics, Suwon-si 16677, Gyeonggi-do, Republic of KoreaDepartment of Electronic Engineering, Sogang University, Seoul 04107, Republic of KoreaDepartment of Artificial Intelligence and Data Science, Sejong University, Seoul 05006, Republic of KoreaRecent advancements in the field of super resolution (SR) have seen the adoption of generative adversarial networks (GANs) for realistic images. In this case, when performing with low-resolution (LR) images, several challenges arise due to the loss of high-frequency details from high-resolution (HR) images, potentially leading to unwanted distortions in the generated SR images. Our paper presents a novel solution by using two-dimensional structure consistency (TSC) for image analysis. The TSC serves as a mask, enabling adaptive analysis based on the unique frequency characteristics of different image regions. Furthermore, a mutual loss mechanism, which dynamically adjusts the training process based on the results filtered by the TSC-based mask, is introduced. Additionally, the TSC loss is proposed to enhance our model capacity to generate precise TSC in high-frequency regions. As a result, our method effectively reduces distortions in high-frequency areas while preserving clarity in regions containing low-frequency components. Our method outperforms other SR techniques, demonstrating superior results in both qualitative and quantitative evaluations. Quantitative measurements, including PSNR, SSIM, and the perceptual metric LPIPS, show comparable PSNR and SSIM values, while the perceptual SR quality is notably improved according to the LPIPS metric.https://www.mdpi.com/2313-433X/10/11/266interpolationimage up-scalingsuper resolutiondeep learning
spellingShingle Yu Lim Seo
Suk-Ju Kang
Yeon-Kug Moon
Frequency Domain-Based Super Resolution Using Two-Dimensional Structure Consistency for Ultra-High-Resolution Display
Journal of Imaging
interpolation
image up-scaling
super resolution
deep learning
title Frequency Domain-Based Super Resolution Using Two-Dimensional Structure Consistency for Ultra-High-Resolution Display
title_full Frequency Domain-Based Super Resolution Using Two-Dimensional Structure Consistency for Ultra-High-Resolution Display
title_fullStr Frequency Domain-Based Super Resolution Using Two-Dimensional Structure Consistency for Ultra-High-Resolution Display
title_full_unstemmed Frequency Domain-Based Super Resolution Using Two-Dimensional Structure Consistency for Ultra-High-Resolution Display
title_short Frequency Domain-Based Super Resolution Using Two-Dimensional Structure Consistency for Ultra-High-Resolution Display
title_sort frequency domain based super resolution using two dimensional structure consistency for ultra high resolution display
topic interpolation
image up-scaling
super resolution
deep learning
url https://www.mdpi.com/2313-433X/10/11/266
work_keys_str_mv AT yulimseo frequencydomainbasedsuperresolutionusingtwodimensionalstructureconsistencyforultrahighresolutiondisplay
AT sukjukang frequencydomainbasedsuperresolutionusingtwodimensionalstructureconsistencyforultrahighresolutiondisplay
AT yeonkugmoon frequencydomainbasedsuperresolutionusingtwodimensionalstructureconsistencyforultrahighresolutiondisplay