Sequential SAR-to-Optical Image Translation

There is a common need for optical sequence images with high spatiotemporal resolution. As a solution, Synthetic Aperture Radar (SAR)-to-optical translation tends to bring high temporal continuity of optical images and low interpretation difficulty of SAR images. Existing studies have been focused o...

Full description

Saved in:
Bibliographic Details
Main Authors: Jingbo Wei, Huan Zhou, Peng Ke, Yaobin Ma, Rongxin Tang
Format: Article
Language:English
Published: MDPI AG 2025-07-01
Series:Remote Sensing
Subjects:
Online Access:https://www.mdpi.com/2072-4292/17/13/2287
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1849319920297312256
author Jingbo Wei
Huan Zhou
Peng Ke
Yaobin Ma
Rongxin Tang
author_facet Jingbo Wei
Huan Zhou
Peng Ke
Yaobin Ma
Rongxin Tang
author_sort Jingbo Wei
collection DOAJ
description There is a common need for optical sequence images with high spatiotemporal resolution. As a solution, Synthetic Aperture Radar (SAR)-to-optical translation tends to bring high temporal continuity of optical images and low interpretation difficulty of SAR images. Existing studies have been focused on converting a single SAR image into a single optical image, failing to utilize the advantages of repeated observations from SAR satellites. To make full use of periodic SAR images, it is proposed to investigate the sequential SAR-to-optical translation, which represents the first effort in this topic. To achieve this, a model based on a diffusion framework has been constructed, with twelve Transformer blocks utilized to effectively capture spatial and temporal features alternatively. A variational autoencoder is employed to encode and decode images, enabling the diffusion model to learn the distribution of features within optical image sequences. A conditional branch is specifically designed for SAR sequences to facilitate feature extraction. Additionally, the capture time is encoded and embedded into the Transformers. Two sequence datasets for the sequence translation task were created, comprising Sentinel-1 Ground Range Detected data and Sentinel-2 red/green/blue data. Our method was tested on new datasets and compared with three state-of-the-art single translation methods. Quantitative and qualitative comparisons validate the effectiveness of the proposed method in maintaining radiometric and spectral consistency.
format Article
id doaj-art-774d90abb23d4543829821b4eccd67a4
institution Kabale University
issn 2072-4292
language English
publishDate 2025-07-01
publisher MDPI AG
record_format Article
series Remote Sensing
spelling doaj-art-774d90abb23d4543829821b4eccd67a42025-08-20T03:50:16ZengMDPI AGRemote Sensing2072-42922025-07-011713228710.3390/rs17132287Sequential SAR-to-Optical Image TranslationJingbo Wei0Huan Zhou1Peng Ke2Yaobin Ma3Rongxin Tang4School of Mathematics and Computer Sciences, Nanchang University, Nanchang 330031, ChinaSchool of Mathematics and Computer Sciences, Nanchang University, Nanchang 330031, ChinaSchool of Information Engineering, Nanchang University, Nanchang 330031, ChinaSchool of Resources and Environment, Nanchang University, Nanchang 330031, ChinaInstitute of Space Science and Technology, Nanchang University, Nanchang 330031, ChinaThere is a common need for optical sequence images with high spatiotemporal resolution. As a solution, Synthetic Aperture Radar (SAR)-to-optical translation tends to bring high temporal continuity of optical images and low interpretation difficulty of SAR images. Existing studies have been focused on converting a single SAR image into a single optical image, failing to utilize the advantages of repeated observations from SAR satellites. To make full use of periodic SAR images, it is proposed to investigate the sequential SAR-to-optical translation, which represents the first effort in this topic. To achieve this, a model based on a diffusion framework has been constructed, with twelve Transformer blocks utilized to effectively capture spatial and temporal features alternatively. A variational autoencoder is employed to encode and decode images, enabling the diffusion model to learn the distribution of features within optical image sequences. A conditional branch is specifically designed for SAR sequences to facilitate feature extraction. Additionally, the capture time is encoded and embedded into the Transformers. Two sequence datasets for the sequence translation task were created, comprising Sentinel-1 Ground Range Detected data and Sentinel-2 red/green/blue data. Our method was tested on new datasets and compared with three state-of-the-art single translation methods. Quantitative and qualitative comparisons validate the effectiveness of the proposed method in maintaining radiometric and spectral consistency.https://www.mdpi.com/2072-4292/17/13/2287SARimage translationsequence translationconditional diffusion model
spellingShingle Jingbo Wei
Huan Zhou
Peng Ke
Yaobin Ma
Rongxin Tang
Sequential SAR-to-Optical Image Translation
Remote Sensing
SAR
image translation
sequence translation
conditional diffusion model
title Sequential SAR-to-Optical Image Translation
title_full Sequential SAR-to-Optical Image Translation
title_fullStr Sequential SAR-to-Optical Image Translation
title_full_unstemmed Sequential SAR-to-Optical Image Translation
title_short Sequential SAR-to-Optical Image Translation
title_sort sequential sar to optical image translation
topic SAR
image translation
sequence translation
conditional diffusion model
url https://www.mdpi.com/2072-4292/17/13/2287
work_keys_str_mv AT jingbowei sequentialsartoopticalimagetranslation
AT huanzhou sequentialsartoopticalimagetranslation
AT pengke sequentialsartoopticalimagetranslation
AT yaobinma sequentialsartoopticalimagetranslation
AT rongxintang sequentialsartoopticalimagetranslation