Adaptive Semantic-Enhanced Denoising Diffusion Probabilistic Model for Remote Sensing Image Super-Resolution

Remote sensing image super-resolution (SR) is a crucial task to restore high-resolution (HR) images from low-resolution (LR) observations. Recently, the denoising diffusion probabilistic model (DDPM) has shown promising performance in image reconstructions by overcoming problems inherent in generati...

Full description

Saved in:
Bibliographic Details
Main Authors: Jialu Sui, Xianping Ma, Xiaokang Zhang, Man-On Pun, Hao Wu
Format: Article
Language:English
Published: IEEE 2025-01-01
Series:IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing
Subjects:
Online Access:https://ieeexplore.ieee.org/document/10763472/
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1846129687775936512
author Jialu Sui
Xianping Ma
Xiaokang Zhang
Man-On Pun
Hao Wu
author_facet Jialu Sui
Xianping Ma
Xiaokang Zhang
Man-On Pun
Hao Wu
author_sort Jialu Sui
collection DOAJ
description Remote sensing image super-resolution (SR) is a crucial task to restore high-resolution (HR) images from low-resolution (LR) observations. Recently, the denoising diffusion probabilistic model (DDPM) has shown promising performance in image reconstructions by overcoming problems inherent in generative models, such as oversmoothing and mode collapse. However, the high-frequency details generated by DDPM often suffer from misalignment with HR images due to the model's tendency to overlook long-range semantic contexts. This challenge is partly due to the prevalent use of a U-Net decoder in the conditional noise predictor, which favors local details and can introduce noise with considerable variance during prediction. To tackle these limitations, an adaptive semantic-enhanced DDPM (ASDDPM) is proposed to enhance the detail-preserving capability of the DDPM by integrating low-frequency semantic insights through a transformer. Specifically, a novel adaptive diffusion transformer decoder is developed to bridge the semantic gap between the encoder and decoder by regulating the noise prediction with the global contextual relationships and long-range dependencies in the diffusion process. In addition, a residual feature fusion strategy establishes information exchange between the two decoders at multiple levels. As a result, the predicted noise generated by our approach closely approximates that of the real noise distribution. Extensive experiments on two SR and two semantic segmentation datasets confirm the superior performance of the proposed ASDDPM in both SR and the subsequent downstream applications.
format Article
id doaj-art-6e919dba6a8e4c5bbc05d1e208a32713
institution Kabale University
issn 1939-1404
2151-1535
language English
publishDate 2025-01-01
publisher IEEE
record_format Article
series IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing
spelling doaj-art-6e919dba6a8e4c5bbc05d1e208a327132024-12-10T00:00:30ZengIEEEIEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing1939-14042151-15352025-01-011889290610.1109/JSTARS.2024.350456910763472Adaptive Semantic-Enhanced Denoising Diffusion Probabilistic Model for Remote Sensing Image Super-ResolutionJialu Sui0https://orcid.org/0000-0002-7450-766XXianping Ma1https://orcid.org/0000-0002-2180-2964Xiaokang Zhang2https://orcid.org/0000-0002-6127-4801Man-On Pun3https://orcid.org/0000-0003-3316-5381Hao Wu4https://orcid.org/0000-0001-5751-7885School of Science and Engineering, The Chinese University of Hong Kong, Shenzhen, ChinaSchool of Science and Engineering, The Chinese University of Hong Kong, Shenzhen, ChinaSchool of Information Science and Engineering, Wuhan University of Science and Technology, Wuhan, ChinaSchool of Science and Engineering, The Chinese University of Hong Kong, Shenzhen, ChinaCollege of Urban and Environment Sciences, Central China Normal University, Wuhan, ChinaRemote sensing image super-resolution (SR) is a crucial task to restore high-resolution (HR) images from low-resolution (LR) observations. Recently, the denoising diffusion probabilistic model (DDPM) has shown promising performance in image reconstructions by overcoming problems inherent in generative models, such as oversmoothing and mode collapse. However, the high-frequency details generated by DDPM often suffer from misalignment with HR images due to the model's tendency to overlook long-range semantic contexts. This challenge is partly due to the prevalent use of a U-Net decoder in the conditional noise predictor, which favors local details and can introduce noise with considerable variance during prediction. To tackle these limitations, an adaptive semantic-enhanced DDPM (ASDDPM) is proposed to enhance the detail-preserving capability of the DDPM by integrating low-frequency semantic insights through a transformer. Specifically, a novel adaptive diffusion transformer decoder is developed to bridge the semantic gap between the encoder and decoder by regulating the noise prediction with the global contextual relationships and long-range dependencies in the diffusion process. In addition, a residual feature fusion strategy establishes information exchange between the two decoders at multiple levels. As a result, the predicted noise generated by our approach closely approximates that of the real noise distribution. Extensive experiments on two SR and two semantic segmentation datasets confirm the superior performance of the proposed ASDDPM in both SR and the subsequent downstream applications.https://ieeexplore.ieee.org/document/10763472/Denoising diffusion probabilistic modelremote sensing imagessingle image super-resolution
spellingShingle Jialu Sui
Xianping Ma
Xiaokang Zhang
Man-On Pun
Hao Wu
Adaptive Semantic-Enhanced Denoising Diffusion Probabilistic Model for Remote Sensing Image Super-Resolution
IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing
Denoising diffusion probabilistic model
remote sensing images
single image super-resolution
title Adaptive Semantic-Enhanced Denoising Diffusion Probabilistic Model for Remote Sensing Image Super-Resolution
title_full Adaptive Semantic-Enhanced Denoising Diffusion Probabilistic Model for Remote Sensing Image Super-Resolution
title_fullStr Adaptive Semantic-Enhanced Denoising Diffusion Probabilistic Model for Remote Sensing Image Super-Resolution
title_full_unstemmed Adaptive Semantic-Enhanced Denoising Diffusion Probabilistic Model for Remote Sensing Image Super-Resolution
title_short Adaptive Semantic-Enhanced Denoising Diffusion Probabilistic Model for Remote Sensing Image Super-Resolution
title_sort adaptive semantic enhanced denoising diffusion probabilistic model for remote sensing image super resolution
topic Denoising diffusion probabilistic model
remote sensing images
single image super-resolution
url https://ieeexplore.ieee.org/document/10763472/
work_keys_str_mv AT jialusui adaptivesemanticenhanceddenoisingdiffusionprobabilisticmodelforremotesensingimagesuperresolution
AT xianpingma adaptivesemanticenhanceddenoisingdiffusionprobabilisticmodelforremotesensingimagesuperresolution
AT xiaokangzhang adaptivesemanticenhanceddenoisingdiffusionprobabilisticmodelforremotesensingimagesuperresolution
AT manonpun adaptivesemanticenhanceddenoisingdiffusionprobabilisticmodelforremotesensingimagesuperresolution
AT haowu adaptivesemanticenhanceddenoisingdiffusionprobabilisticmodelforremotesensingimagesuperresolution