1.西南交通大学地球科学与工程学院,四川 成都,611756
2.高速铁路安全运营空间信息技术国家地方联合工程实验室,四川 成都,611756
吴仁哲,博士,主要从事遥感智能解译及冰川冰湖动态演化研究。rswrz@hnas.ac.cn
刘国祥,博士,教授。rsgxliu@swjtu.edu.cn
收稿:2024-04-16,
纸质出版:2026-02-05
移动端阅览
吴仁哲,刘国祥,张瑞,等.一种适用于高山复杂气候条件下的加权掩膜云去除模型[J].武汉大学学报(信息科学版),2026,51(2):357-366.
WU Renzhe,LIU Guoxiang,ZHANG Rui,et al.A Weighted Masking Cloud Removal Model Suitable for Complex Alpine Climatic Conditions[J].Geomatics and Information Science of Wuhan University,2026,51(2):357-366.
吴仁哲,刘国祥,张瑞,等.一种适用于高山复杂气候条件下的加权掩膜云去除模型[J].武汉大学学报(信息科学版),2026,51(2):357-366. DOI:10.13203/j.whugis20240020
WU Renzhe,LIU Guoxiang,ZHANG Rui,et al.A Weighted Masking Cloud Removal Model Suitable for Complex Alpine Climatic Conditions[J].Geomatics and Information Science of Wuhan University,2026,51(2):357-366. DOI:10.13203/j.whugis20240020
云雾覆盖是制约光学遥感研究的关键因素之一,尤其是在藏东南地区,其平均云量高达62%,严重降低了光学遥感影像的可用性。与时序去云方法相比,单幅影像去云具有更强的时效性,但实现难度较高。且现有的云层不透度估计方法难以恢复厚云覆盖区域的地表信息,而生成对抗网络方法则常常伴随伪影且可解释性差。针对上述问题,提出了一种适用于高山复杂气候条件下的加权掩膜云去除模型。所提模型采用改进的Transformer模块作为生成器,并重构了多头注意力机制,有效融合了薄云去除与厚云地表信息修复的能力。经哨兵2号Level-2A数据测试表明,所提模型在高保真恢复地表信息的同时有效减少了厚云修复时的伪影现象。实验验证结果表明,所提模型在各项性能指标上均优于现有方法,实现了高保真度的地表信息恢复,其平均绝对误差和均方根误差分别为0.025 6和0.035 6,峰值信噪比和结构相似性则分别为30.185 1和0.899 6。该研究成果可为高山复杂气候条件下的云雾去除提供参考,并为山地研究提供数据支持。
Objectives
2
Cloud cover hampers optical remote sensing
especially in high-altitude alpine regions. In Southeastern Tibetan Plateau
cloud cover reaches 62%
which limits the use of optical image. While single-image cloud removal method is faster than temporal methods
it faces technical hurdles. The existing methods struggle under thick clouds
and generative adversarial network-based approaches often produce artifacts and poor interpretability. We propose a weighted masking cloud removal model designed for complex alpine climates
aiming for accurate surface restoration under various cloud conditions
reducing artifacts and improving robustness in mountainous terrains with snow cover.
Methods
2
The proposed model combines cloud opacity estimation with an advanced image generation framework. It begins by formalizing assumptions about cloud opacity and brightness to address stratified cloud phenomena in remote sensing images. And its core is an improved Transformer-based generator
which introduces a redesigned multi-head mask attention (MMA) mechanism. This mechanism uses a dynamically generated weighted mask created from estimated cloud opacity and brightness compensation maps to modulate neuron activation. The mask strategically suppresses features from heavily cloud-obscured pixels during aggregation
focusing the model's attention first on clearer regions and cloud edges. Additionally
a progressive sliding window mask update strategy is adopted
which gradually shrinks the inhibitory mask as the network deepens. This allows the model to iteratively propagate reliable information from outer regions into thick cloud cores
enabling full restoration. The model architecture adopts a dual-generator design
in which the first generator is responsible for estimating cloud opacity and brightness compensation values to generate guiding masks and the second generator integrated with MMA blocks is dedicated to synthesizing the final cloud-free image. Training is guided by a composite loss function that integrates non-saturating adversarial loss
structural similarity (SSIM) loss
and perceptual loss
ensuring visual fidelity
structural accuracy
and feature consistency. For evaluation
a dedicated alpine dataset is built using Sentinel-2 Level-2A imagery in Southeastern Tibetan Plateau. Cloud masks from S2cloudless are used to create precise cloudy-clear pairs
and a simulation protocol generates physically realistic training samples with corresponding opacity and compensation ground truth.
Results
2
The proposed model is rigorously evaluated against three state-of-the-art methods. On quantitative metrics
it outperforms all counterparts
achieving a mean absolute error of 0.025 6
root mean square error of 0.035 6
peak signal-to-noise ratio of 30.185 1
and SSIM index of 0.899 6. Visually
it excels in texture preservation and artifact reduction
especially under moderate (20%⁃30%) and heavy (>30%) cloud cover
where others show significant distortion. The proposed model also demonstrates robustness in snow-covered alpine scenes.
Conclusions
2
This work presents an effective cloud removal model for alpine regions by unifying thin-cloud correction and thick-cloud inpainting within a Transformer framework. The weighted mask and progressive update strategy provide a targeted explainable restoration process. This approach significantly enhances optical image usability in cloudy mountain areas and offers valuable support for related scientific research.
王志林 , 冉盈盈 . 遥感影像去除云阴影的高山区冰湖信息提取方法 [J]. 北京测绘 , 2019 , 33 ( 10 ): 1236 - 1239 .
WANG Zhilin , RAN Yinging . A Cloud Shadow Removel Method for the Extraction of High Mountain Lake [J]. Beijing Surveying and Mapping , 2019 , 33 ( 10 ): 1236 - 1239 .
苗顺霞 , 孙开敏 , 胡秀清 , 等 . 基于风云三号D星MERSI-II影像的青藏高原湖泊范围监测能力分析 [J]. 武汉大学学报(信息科学版) , 2025 , 50 ( 12 ): 2526 - 2536 .
MIAO Shunxia , SUN Kaimin , HU Xiuqing , et al . Capability Analysis of Lake Extent Monitoring in the Tibetan Plateau from MERSI-II Images of FY-3D Satellite [J]. Geomatics and Information Science of Wuhan University , 2025 , 50 ( 12 ): 2526 - 2536 .
戴可人 , 许强 , 邓晋 , 等 . 滑坡隐患InSAR广域识别研究综述 [J]. 武汉大学学报(信息科学版) , 2025 , 50 ( 8 ): 1469 - 1482 .
DAI Keren , XU Qiang , DENG Jin , et al . Review on Large-Scale Identification of Potential Landslides with InSAR [J]. Geomatics and Information Science of Wuhan University , 2025 , 50 ( 8 ): 1469 - 1482 .
ROSSOW W B , SCHIFFER R A . Advances in Understanding Clouds from ISCCP [J]. Bulletin of the American Meteorological Society , 1999 , 80 ( 11 ): 2261 - 2287 .
ZHANG Y C , ROSSOW W B , LACIS A A , et al . Calculation of Radiative Fluxes from the Surface to Top of Atmosphere Based on ISCCP and Other Global Data Sets: Refinements of the Radiative Transfer Model and the Input Data [J]. Journal of Geophysical Research: Atmospheres , 2004 , 109 ( D19 ): 2003 JD004457.
陈奋 , 闫冬梅 , 赵忠明 . 基于无抽样小波的遥感影像薄云检测与去除 [J]. 武汉大学学报(信息科学版) , 2007 , 32 ( 1 ): 71 - 74 .
CHEN Fen , YAN Dongmei , ZHAO Zhongming . Haze Detection and Removal in Remote Sensing Images Based on Undecimated Wavelet Transform [J]. Geomatics and Information Science of Wuhan University , 2007 , 32 ( 1 ): 71 - 74 .
贺辉 , 彭望琭 , 匡锦瑜 . 自适应滤波的高分辨率遥感影像薄云去除算法 [J]. 地球信息科学学报 , 2009 , 11 ( 3 ): 305 - 311 .
HE Hui , PENG Wanglu , KUANG Jinyu . Thin Cloud Cover Removed from High‐Resolution Remote Sensing Images Based on the Adaptive Filtering and Gray‐Scale Transformation [J]. Journal of Geo‐Information Science , 2009 , 11 ( 3 ): 305 - 311 .
ZHU X L , GAO F , LIU D S , et al . A Modified Neighborhood Similar Pixel Interpolator Approach for Removing Thick Clouds in Landsat Images [J]. IEEE Geoscience and Remote Sensing Letters , 2012 , 9 ( 3 ): 521 - 525 .
谢华美 , 李荣艳 , 田艳琴 , 等 . 基于大数据量遥感图像的薄云去除 [J]. 北京师范大学学报(自然科学版) , 2006 , 42 ( 1 ): 42 - 46 .
XIE Huamei , LI Rongyan , TIAN Yanqin , et al . The Removing Clouds Method Based on Large Re‐mote Sensing Image [J]. Journal of Beijing Normal University (Natural Science) , 2006 , 42 ( 1 ): 42 - 46 .
董彦芳 , 孙国清 , 庞勇 , 等 . 几种用于 TM 图像薄云去除的大气纠正算法比较 [J]. 国土资源遥感 , 2005 , 17 ( 2 ): 16 - 19 .
DONG Yanfang , SUN Guoqing , PANG Yong , et al . A Comparitive Study of Some Atmospheric Cor‐rection Methods [J]. Remote Sensing for Land & Re‐sources , 2005 , 17 ( 2 ): 16 - 19 .
XU M , DENG F R , JIA S , et al . Attention Mechanism‐Based Generative Adversarial Networks for Cloud Removal in Landsat Images [J]. Remote Sensing of Environment , 2022 , 271 : 112902 .
张雨姝 , 戴佩玉 . 一种边缘辅助的卫星影像云修复卷积神经网络 [J]. 测绘地理信息 , 2024 , 49 ( 2 ): 81 - 86 .
ZHANG Yushu , DAI Peiyu . Cloud Removal of Satellite Image Using Edge-Guided Convolutional Neural Network [J]. Journal of Geomatics , 2024 , 49 ( 2 ): 81 - 86 .
MA D Y , WU R Z , XIAO D S , et al . Cloud Removal from Satellite Images Using a Deep Learning Model with the Cloud‑Matting Method [J]. Remote Sensing , 2023 , 15 ( 4 ): 904 .
WU R Z , LIU G X , LV J C , et al . An Innovative Approach for Effective Removal of Thin Clouds in Optical Images Using Convolutional Matting Model [J]. Remote Sensing , 2023 , 15 ( 8 ): 2119 .
戴佩玉 , 李世忠 , 季顺平 , 等 . 一种基于域自适应泛化增强的云检测方法 [J]. 武汉大学学报(信息科学版) , 2025 , 50 ( 1 ): 110 - 119 .
CUI Peng , DAI Peiyu , LI Shizhong , JI Shunping , et al . A Cloud Detection Method with Domain Adaptation Enhanced Generalization Capability [J]. Geomatics and Information Science of Wuhan University , 2025 , 50 ( 1 ): 110 - 119 .
崔鹏 , 郭晓军 , 姜天海 , 等 . “亚洲水塔”变化的灾害效应与减灾对策 [J]. 中国科学院院刊 , 2019 , 34 ( 11 ): 1313 - 1321 .
CUI Peng , GUO Xiaojun , JIANG Tianhai , et al . Disaster Effect Induced by Asian Water Tower Change and Mitigation Strategies [J]. Bulletin of Chinese Academy of Sciences , 2019 , 34 ( 11 ): 1313 - 1321 .
SKAKUN S , WEVERS J , BROCKMANN C , et al . Cloud Mask Intercomparison eXercise (CMIX): An Evaluation of Cloud Masking Algorithms for Landsat 8 and Sentinel‑2 [J]. Remote Sensing of Environment , 2022 , 274 : 112990 .
YANG K , WU H , QIN J , et al . Recent Climate Changes over the Tibetan Plateau and Their Impacts on Energy and Water Cycle: A Review [J]. Global and Planetary Change , 2014 , 112 : 79 - 91 .
刘国祥 , 张波 , 张瑞 , 等 . 联合卫星SAR和地基SAR的海螺沟冰川动态变化及次生滑坡灾害监测 [J]. 武汉大学学报(信息科学版) , 2019 , 44 ( 7 ): 980 - 995 .
LIU Guoxiang , ZHANG Bo , ZHANG Rui , et al . Monitoring Dynamics of Hailuogou Glacier and the Secondary Landslide Disasters Based on Combination of Satellite SAR and Ground‑Based SAR [J]. Geomatics and Information Science of Wuhan University , 2019 , 44 ( 7 ): 980 - 995 .
MAUSSION F , SCHERER D , MÖLG T , et al . Precipitation Seasonality and Variability over the Tibetan Plateau as Resolved by the High Asia Reanalysis [J]. Journal of Climate , 2014 , 27 ( 5 ): 1910 - 1927 .
杜建括 , 何元庆 , 李双 , 等 . 横断山区典型海洋型冰川物质平衡研究 [J]. 地理学报 , 2015 , 70 ( 9 ): 1415 - 1422 .
DU Jiankuo , HE Yuanqing , LI Shuang , et al . Mass Balance of a Typical Monsoonal Temperate Glacier in Hengduan Mountains Region [J]. Acta Geographica Sinica , 2015 , 70 ( 9 ): 1415 - 1422 .
刘时银 , 姚晓军 , 郭万钦 , 等 . 基于第二次冰川编目的中国冰川现状 [J]. 地理学报 , 2015 , 70 ( 1 ): 3 - 16 .
LIU Shiyin , YAO Xiaojun , GUO Wanqin , et al . The Contemporary Glaciers in China Based on the Second Chinese Glacier Inventory [J]. Acta Geographica Sinica , 2015 , 70 ( 1 ): 3 - 16 .
ZHOU X , LIN D Y , LIU J Y . RICE: A Dataset and Baseline for Cloud Removal in Remote Sensing Images [J]. Journal of Combinatorial Mathematics and Combinatorial Computing , 2024 , 120 : 107 - 124 .
LI J , WU Z C , HU Z W , et al . Deep Learning Based Thin Cloud Removal Fusing Vegetation Red Edge and Short Wave Infrared Spectral Information for Sentinel‑2A Imagery [J]. Remote Sensing , 2021 , 13 ( 1 ): 157 .
EBEL P , XU Y J , SCHMITT M , et al . SEN12MS‑CR‑TS: A Remote‑Sensing Data Set for Multimodal Multitemporal Cloud Removal [J]. IEEE Transactions on Geoscience and Remote Sensing , 2022 , 60 : 5222414 .
LONG C J , LI X H , JING Y H , et al . Bishift Networks for Thick Cloud Removal with Multitemporal Remote Sensing Images [J]. International Journal of Intelligent Systems , 2023 , 2023 ( 1 ): e9953198 .
LI J , WU Z C , HU Z W , et al . Thin Cloud Removal in Optical Remote Sensing Images Based on Generative Adversarial Networks and Physical Model of Cloud Distortion [J]. ISPRS Journal of Photogrammetry and Remote Sensing , 2020 , 166 : 373 - 389 .
WANG Z , BOVIK A C , SHEIKH H R , et al . Image Quality Assessment: From Error Visibility to Structural Similarity [J]. IEEE Transactions on Image Processing , 2004 , 13 ( 4 ): 600 - 612 .
CHEN Y Y , CAI Z C , YUAN J Y , et al . A Novel Dense‑Attention Network for Thick Cloud Removal by Reconstructing Semantic Information [J]. IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing , 2023 , 16 : 2339 - 2351 .
EMAMI H , ALIABADI M M , DONG M , et al . SPA‑GAN: Spatial Attention GAN for Image‐to‐Image Translation [J]. IEEE Transactions on Multimedia , 2021 , 23 : 391 - 401 .
LI Y S , WEI F Y , ZHANG Y J , et al . HS 2 P: Hierarchical Spectral and Structure‑Preserving Fusion Network for Multimodal Remote Sensing Image Cloud and Shadow Removal [J]. Information Fusion , 2023 , 94 : 215 - 228 .
0
浏览量
1
下载量
0
CSCD
关联资源
相关文章
相关作者
相关机构
京公网安备11010802024621