pith. sign in

arxiv: 2510.07905 · v4 · submitted 2025-10-09 · 📡 eess.IV · cs.CV· cs.MM

SatFusion: A Unified Framework for Enhancing Remote Sensing Images via Multi-Frame and Multi-Source Images Fusion

Yufei Tong , Guanjie Cheng , Peihan Wu , Feiyi Chen , Xinkui Zhao , Shuiguang Deng This is my paper

Pith reviewed 2026-05-18 09:01 UTC · model grok-4.3

classification 📡 eess.IV cs.CVcs.MM
keywords remote sensingimage fusionmulti-frame super-resolutionpansharpeningmulti-source fusiontransformer aggregationimage enhancement
0
0 comments X

The pith

Coupling multi-frame and multi-source priors improves remote sensing image reconstruction fidelity and robustness.

A machine-rendered reading of the paper's core claim, the machinery that carries it, and where it could break.

The paper introduces SatFusion as a unified framework that fuses multiple low-resolution multispectral frames with a high-resolution panchromatic image. It claims existing approaches suffer when studied in isolation because multi-frame methods lack fine structural priors while pansharpening methods depend on upsampled inputs and falter under noise or misalignment. SatFusion addresses this by using one module to aggregate semantic features across frames and another to inject panchromatic details with implicit alignment. An advanced variant adds panchromatic guidance during the frame-fusion stage through structure-aware embedding and adaptive transformer aggregation. A sympathetic reader would care because the result is higher-quality reconstructions that support more reliable downstream analysis in remote sensing without requiring new sensors.

Core claim

SatFusion extracts high-resolution semantic features by aggregating complementary information from multiple low-resolution multispectral frames via the Multi-Frame Image Fusion module and integrates fine-grained structural details from a high-resolution panchromatic image through the Multi-Source Image Fusion module with implicit pixel-level alignment. The advanced SatFusion* variant incorporates a panchromatic-guided mechanism into the multi-frame stage, employing structure-aware feature embedding and transformer-based adaptive aggregation to enable spatially adaptive feature selection and strengthen coupling between the two representations.

What carries the argument

The Multi-Frame Image Fusion and Multi-Source Image Fusion modules that perform complementary aggregation from low-resolution multispectral frames and structural integration from a high-resolution panchromatic image respectively, with the transformer-based adaptive aggregation in the advanced variant.

If this is right

  • Superior reconstruction fidelity across four benchmark remote sensing datasets.
  • Greater robustness to noise and misalignment than isolated multi-frame or pansharpening methods.
  • Improved generalizability when multi-frame and multi-source priors are coupled.
  • Spatially adaptive feature selection that strengthens representation coupling in the guided variant.

Where Pith is reading between the lines

These are editorial extensions of the paper, not claims the author makes directly.

  • The implicit alignment approach may reduce reliance on explicit co-registration preprocessing steps in operational pipelines.
  • The same coupling principle could be tested in other multi-modal imaging settings where frame sequences and a higher-resolution reference channel are both available.
  • Computational scaling experiments on larger satellite scenes would clarify whether the transformer aggregation remains practical for onboard processing.

Load-bearing premise

The modules can aggregate information from low-resolution multispectral frames and the high-resolution panchromatic image through implicit alignment without creating artifacts from noise or misalignment.

What would settle it

A benchmark dataset containing remote sensing images with documented misalignment or elevated noise levels where SatFusion produces lower quality metrics or visible artifacts compared with applying multi-frame super-resolution or pansharpening separately.

Figures

Figures reproduced from arXiv: 2510.07905 by Feiyi Chen, Guanjie Cheng, Peihan Wu, Shuiguang Deng, Xinkui Zhao, Yufei Tong.

Figure 1
Figure 1. Figure 1: Overview of the SatFusion system. LEO satellites transmit low-quality images to ground stations, where SatFu￾sion performs fusion-based reconstruction to enhance image quality and reduce redundancy. images, which makes it highly sensitive to input quality and thus difficult to apply effectively in practice. To address the aforementioned background and challenges, we propose SatFusion, a novel unified frame… view at source ↗
Figure 2
Figure 2. Figure 2: Overview of MISR and Pansharpening network [PITH_FULL_IMAGE:figures/full_fig_p003_2.png] view at source ↗
Figure 3
Figure 3. Figure 3: Architecture of SatFusion. The framework takes multiple multi-temporal LRMS images and a single HRPAN image as inputs. These are processed sequentially through the multi-temporal image fusion module, the multi-source image fusion module, and the fusion composition module, producing the final HRMS image. where 𝑅𝑒𝑠𝑖𝑧𝑒 adjusts the spatial dimensions by interpolation to ensure that the output features of the M… view at source ↗
Figure 4
Figure 4. Figure 4: (left) Workflow of the conventional Wald protocol [PITH_FULL_IMAGE:figures/full_fig_p006_4.png] view at source ↗
Figure 5
Figure 5. Figure 5: Quantitative comparison of PSNR for the fused images on the WorldStrat real-world dataset. [PITH_FULL_IMAGE:figures/full_fig_p008_5.png] view at source ↗
Figure 6
Figure 6. Figure 6: Qualitative comparison of error maps between the fused and ground-truth images. [PITH_FULL_IMAGE:figures/full_fig_p008_6.png] view at source ↗
Figure 7
Figure 7. Figure 7: Difference in data volume between input and output [PITH_FULL_IMAGE:figures/full_fig_p009_7.png] view at source ↗
Figure 8
Figure 8. Figure 8: Changes in metrics for different fusion methods as [PITH_FULL_IMAGE:figures/full_fig_p010_8.png] view at source ↗
Figure 9
Figure 9. Figure 9: PSNR comparison under different perturbation lev [PITH_FULL_IMAGE:figures/full_fig_p012_9.png] view at source ↗
read the original abstract

High-quality remote sensing (RS) image acquisition is fundamentally constrained by physical limitations. While Multi-Frame Super-Resolution (MFSR) and Pansharpening address this by exploiting complementary information, they are typically studied in isolation: MFSR lacks high-resolution (HR) structural priors for fine-grained texture recovery, whereas Pansharpening relies on upsampled low-resolution (LR) inputs and is sensitive to noise and misalignment. In this paper, we propose SatFusion, a novel and unified framework that seamlessly bridges multi-frame and multi-source RS image fusion. SatFusion extracts HR semantic features by aggregating complementary information from multiple LR multispectral frames via a Multi-Frame Image Fusion (MFIF) module, and integrates fine-grained structural details from an HR panchromatic image through a Multi-Source Image Fusion (MSIF) module with implicit pixel-level alignment. To further alleviate the lack of structural priors during multi-frame fusion, we introduce an advanced variant, SatFusion*, which integrates a panchromatic-guided mechanism into the MFIF stage. Through structure-aware feature embedding and transformer-based adaptive aggregation, SatFusion* enables spatially adaptive feature selection, strengthening the coupling between multi-frame and multi-source representations. Extensive experiments on four benchmark datasets validate our core insight: synergistically coupling multi-frame and multi-source priors effectively resolves the fragility of existing paradigms, delivering superior reconstruction fidelity, robustness, and generalizability.

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, simulated authors' rebuttal, and a circularity audit. Tearing a paper down is the easy half of reading it; the pith above is the substance, this is the friction.

Referee Report

2 major / 2 minor

Summary. The paper proposes SatFusion, a unified framework for remote sensing image enhancement that bridges Multi-Frame Super-Resolution (MFSR) and Pansharpening. It introduces an MFIF module to aggregate complementary information from multiple low-resolution multispectral frames and an MSIF module to integrate fine-grained structural details from a high-resolution panchromatic image via implicit pixel-level alignment. An advanced variant, SatFusion*, incorporates a panchromatic-guided mechanism into MFIF using structure-aware feature embedding and transformer-based adaptive aggregation. The authors claim that this synergistic coupling resolves limitations of existing isolated paradigms and delivers superior reconstruction fidelity, robustness, and generalizability, validated through experiments on four benchmark datasets.

Significance. If the empirical claims are substantiated with quantitative evidence, the work could meaningfully advance remote sensing image fusion by providing a single architecture that exploits both temporal/multi-frame and cross-modal (multispectral-panchromatic) priors. The transformer-based adaptive aggregation and implicit alignment mechanisms offer a potentially generalizable alternative to separate MFSR or pansharpening pipelines, which is relevant for applications requiring high-fidelity satellite imagery under varying acquisition conditions.

major comments (2)
  1. [Abstract] Abstract: the central claims of 'superior reconstruction fidelity, robustness, and generalizability' are asserted without any quantitative metrics, error bars, ablation results, or details on how misalignment and noise were handled in the four benchmark datasets, leaving the load-bearing empirical support for the synergistic coupling unverified.
  2. [Method (MFIF/MSIF)] MFIF and MSIF module descriptions: the implicit pixel-level alignment in MSIF and the structure-aware embedding in SatFusion* are presented without an explicit registration step, without a misalignment simulation protocol (e.g., sub-pixel to multi-pixel shifts or rotations typical in satellite data), and without ablations on realistic RS noise/misalignment magnitudes; if the implicit mechanism fails outside benchmark alignment statistics, the claimed advantage over isolated MFSR or pansharpening collapses.
minor comments (2)
  1. [Method] Notation for the transformer-based aggregation could be clarified with explicit equations for the adaptive feature selection weights.
  2. [Introduction] The distinction between SatFusion and SatFusion* should be summarized in a table or diagram for quick comparison.

Simulated Author's Rebuttal

2 responses · 0 unresolved

We thank the referee for the constructive and detailed feedback on our manuscript. We appreciate the opportunity to address the concerns raised and clarify the contributions of SatFusion. Below we respond point-by-point to the major comments.

read point-by-point responses

  1. Referee: [Abstract] Abstract: the central claims of 'superior reconstruction fidelity, robustness, and generalizability' are asserted without any quantitative metrics, error bars, ablation results, or details on how misalignment and noise were handled in the four benchmark datasets, leaving the load-bearing empirical support for the synergistic coupling unverified.

    Authors: We agree that the abstract would be strengthened by including concrete quantitative support for the central claims. In the revised version we will incorporate key metrics (e.g., average PSNR/SSIM gains and robustness indicators) drawn from the experimental results on the four benchmark datasets. Full quantitative tables, error bars from repeated runs, and ablation studies already appear in Sections 4 and 5; we will add a brief reference to these sections in the abstract. Dataset preprocessing, including standard handling of noise and alignment statistics for the chosen benchmarks, is described in Section 3.2 and will be cross-referenced. revision: yes

  2. Referee: [Method (MFIF/MSIF)] MFIF and MSIF module descriptions: the implicit pixel-level alignment in MSIF and the structure-aware embedding in SatFusion* are presented without an explicit registration step, without a misalignment simulation protocol (e.g., sub-pixel to multi-pixel shifts or rotations typical in satellite data), and without ablations on realistic RS noise/misalignment magnitudes; if the implicit mechanism fails outside benchmark alignment statistics, the claimed advantage over isolated MFSR or pansharpening collapses.

    Authors: The MSIF module is deliberately designed around implicit pixel-level alignment via cross-modal feature correlation and attention, avoiding an explicit registration step that can introduce artifacts in satellite imagery; this rationale is stated in Section 3.3. We acknowledge, however, that additional empirical validation would be beneficial. In the revision we will add a dedicated robustness subsection that includes a misalignment simulation protocol (sub-pixel to multi-pixel shifts and rotations) and ablations across realistic RS noise magnitudes, demonstrating that the implicit mechanism maintains performance beyond the alignment statistics of the original benchmarks. revision: yes

Circularity Check

0 steps flagged

No circularity in SatFusion architectural proposal

full rationale

The paper introduces SatFusion as a new unified framework combining MFIF and MSIF modules via structure-aware embedding and transformer-based aggregation, with an optional panchromatic-guided variant in SatFusion*. Claims rest on the design of these modules and experimental results across four benchmark datasets rather than any mathematical derivation, parameter fitting, or self-citation chain that reduces to the inputs by construction. No equations or steps match the enumerated circularity patterns; the central insight is presented as an empirical architectural combination.

Axiom & Free-Parameter Ledger

0 free parameters · 0 axioms · 0 invented entities

Abstract-only review provides no explicit free parameters, axioms, or invented entities; full text would be needed to audit modeling choices such as alignment assumptions or feature embedding details.

pith-pipeline@v0.9.0 · 5804 in / 1179 out tokens · 30905 ms · 2026-05-18T09:01:02.543200+00:00 · methodology

discussion (0)

Sign in with ORCID, Apple, or X to comment. Anyone can read and Pith papers without signing in.

Lean theorems connected to this paper

Citations machine-checked in the Pith Canon. Every link opens the source theorem in the public Lean library.

What do these tags mean?
matches
The paper's claim is directly supported by a theorem in the formal canon.
supports
The theorem supports part of the paper's argument, but the paper may add assumptions or extra steps.
extends
The paper goes beyond the formal theorem; the theorem is a base layer rather than the whole result.
uses
The paper appears to rely on the theorem as machinery.
contradicts
The paper's claim conflicts with a theorem or certificate in the canon.
unclear
Pith found a possible connection, but the passage is too broad, indirect, or ambiguous to say the theorem truly supports the claim.

Reference graph

Works this paper leans on

54 extracted references · 54 canonical work pages · 1 internal anchor

  1. [1]

    Luciano Alparone, Lucien Wald, Jocelyn Chanussot, Claire Thomas, Paolo Gamba, and Lori Mann Bruce. 2007. Comparison of pansharpening algorithms: Outcome of the 2006 GRS-S data-fusion contest.IEEE Transactions on Geoscience and Remote Sensing45, 10 (2007), 3012–3021

    work page 2007

  2. [2]

    Tai An, Xin Zhang, Chunlei Huo, Bin Xue, Lingfeng Wang, and Chunhong Pan. 2022. TR-MISR: Multiimage super-resolution based on feature fusion with transformers.IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing15 (2022), 1373–1388

    work page 2022

  3. [3]

    Wele Gedara Chaminda Bandara and Vishal M Patel. 2022. Hypertransformer: A textural and spectral feature fusion transformer for pansharpening. InProceedings of the IEEE/CVF conference on computer vision and pattern recognition. 1767–1777

    work page 2022

  4. [4]

    Matteo Ciotola, Sergio Vitale, Antonio Mazza, Giovanni Poggi, and Giuseppe Scarpa. 2022. Pansharpening by convolutional neural networks in the full resolu- tion framework.IEEE Transactions on Geoscience and Remote Sensing60 (2022), 1–17

    work page 2022

  5. [5]

    Julien Cornebise, Ivan Oršolić, and Freddie Kalaitzis. 2022. Open high-resolution satellite imagery: The worldstrat dataset–with application to super-resolution. Advances in Neural Information Processing Systems35 (2022), 25979–25991

    work page 2022

  6. [6]

    Bradley Denby and Brandon Lucia. 2020. Orbital edge computing: Nanosatellite constellations as a new class of computer system. InProceedings of the Twenty- Fifth International Conference on Architectural Support for Programming Languages and Operating Systems. 939–954

    work page 2020

  7. [7]

    Liang-Jian Deng, Gemine Vivone, Cheng Jin, and Jocelyn Chanussot. 2020. Detail injection-based deep convolutional neural networks for pansharpening.IEEE Transactions on Geoscience and Remote Sensing59, 8 (2020), 6995–7010

    work page 2020

  8. [8]

    Liang-Jian Deng, Gemine Vivone, Mercedes E Paoletti, Giuseppe Scarpa, Jiang He, Yongjun Zhang, Jocelyn Chanussot, and Antonio Plaza. 2022. Machine learning in pansharpening: A benchmark, from shallow to deep networks.IEEE Geoscience and Remote Sensing Magazine10, 3 (2022), 279–315

    work page 2022

  9. [9]

    Michel Deudon, Alfredo Kalaitzis, Israel Goytom, Md Rifat Arefin, Zhichao Lin, Kris Sankaran, Vincent Michalski, Samira E Kahou, Julien Cornebise, and Yoshua Bengio. 2020. Highres-net: Recursive fusion for multi-frame super-resolution of satellite imagery.arXiv preprint arXiv:2002.06460(2020)

    work page arXiv 2020

  10. [10]

    Chao Dong, Chen Change Loy, Kaiming He, and Xiaoou Tang. 2014. Learning a deep convolutional network for image super-resolution. InComputer Vision– ECCV 2014: 13th European Conference, Zurich, Switzerland, September 6-12, 2014, Proceedings, Part IV 13. Springer, 184–199

    work page 2014

  11. [11]

    Alexey Dosovitskiy, Lucas Beyer, Alexander Kolesnikov, Dirk Weissenborn, Xi- aohua Zhai, Thomas Unterthiner, Mostafa Dehghani, Matthias Minderer, Georg Heigold, Sylvain Gelly, et al. 2020. An image is worth 16x16 words: Transformers for image recognition at scale.arXiv preprint arXiv:2010.11929(2020)

    work page internal anchor Pith review Pith/arXiv arXiv 2020

  12. [12]

    Yule Duan, Xiao Wu, Haoyu Deng, and Liang-Jian Deng. 2024. Content-adaptive non-local convolution for remote sensing pansharpening. InProceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition. 27738–27747

    work page 2024

  13. [13]

    Jing Gao, Peng Li, Zhikui Chen, and Jianing Zhang. 2020. A survey on deep learning for multimodal data fusion.Neural Computation32, 5 (2020), 829–864

    work page 2020

  14. [14]

    Lin He, Yizhou Rao, Jun Li, Jocelyn Chanussot, Antonio Plaza, Jiawei Zhu, and Bo Li. 2019. Pansharpening via detail injection based convolutional neural networks. IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing 12, 4 (2019), 1188–1204

    work page 2019

  15. [15]

    Xuanhua He, Ke Cao, Jie Zhang, Keyu Yan, Yingying Wang, Rui Li, Chengjun Xie, Danfeng Hong, and Man Zhou. 2025. Pan-mamba: Effective pan-sharpening with state space model.Information Fusion115 (2025), 102779

    work page 2025

  16. [16]

    Sergio Herrería-Alonso, Miguel Rodríguez-Pérez, Raúl F Rodríguez-Rubio, and Fernando Pérez-Fontán. 2023. Improving uplink scalability of LoRa-based direct- to-satellite IoT networks.IEEE Internet of Things Journal11, 7 (2023), 12526– 12535

    work page 2023

  17. [17]

    Thorsten Hoeser, Felix Bachofer, and Claudia Kuenzer. 2020. Object detection and image segmentation with deep learning on Earth observation data: A review—Part II: Applications.Remote Sensing12, 18 (2020), 3053

    work page 2020

  18. [18]

    Thorsten Hoeser and Claudia Kuenzer. 2020. Object detection and image segmen- tation with deep learning on earth observation data: A review-part i: Evolution and recent trends.Remote Sensing12, 10 (2020), 1667

    work page 2020

  19. [19]

    Jacob Høxbroe Jeppesen, Rune Hylsberg Jacobsen, Fadil Inceoglu, and Thomas Skjødeberg Toftegaard. 2019. A cloud detection algorithm for satel- lite imagery based on deep learning.Remote sensing of environment229 (2019), 247–259

    work page 2019

  20. [20]

    Zi-Rong Jin, Tian-Jing Zhang, Tai-Xiang Jiang, Gemine Vivone, and Liang-Jian Deng. 2022. LAGConv: Local-context adaptive convolution kernels with global harmonic bias for pansharpening. InProceedings of the AAAI conference on artifi- cial intelligence, Vol. 36. 1113–1121

    work page 2022

  21. [21]

    Taeyeoun Kim, Jeongho Kwak, and Jihwan P Choi. 2021. Satellite edge computing architecture and network slice scheduling for IoT support.IEEE Internet of Things journal9, 16 (2021), 14938–14951

    work page 2021

  22. [22]

    Vivek Kothari, Edgar Liberis, and Nicholas D Lane. 2020. The final frontier: Deep learning in space. InProceedings of the 21st international workshop on mobile computing systems and applications. 45–49

    work page 2020

  23. [23]

    Israel Leyva-Mayorga, Marc Martinez-Gost, Marco Moretti, Ana Pérez-Neira, Miguel Ángel Vázquez, Petar Popovski, and Beatriz Soret. 2023. Satellite edge Conference acronym ’XX, June 03–05, 2018, Woodstock, NY Tong et al. computing for real-time and very-high resolution earth observation.IEEE Trans- actions on Communications71, 10 (2023), 6180–6194

    work page 2023

  24. [24]

    Jiaxin Li, Danfeng Hong, Lianru Gao, Jing Yao, Ke Zheng, Bing Zhang, and Jocelyn Chanussot. 2022. Deep learning in multimodal remote sensing data fusion: A comprehensive review.International Journal of Applied Earth Observation and Geoinformation112 (2022), 102926

    work page 2022

  25. [25]

    Qing Li and Daliang Xu. 2023. Satellite Computing: From Space to Your Screen. InInternational Conference on Service-Oriented Computing. Springer, 343–349

    work page 2023

  26. [26]

    Ze Liu, Yutong Lin, Yue Cao, Han Hu, Yixuan Wei, Zheng Zhang, Stephen Lin, and Baining Guo. 2021. Swin transformer: Hierarchical vision transformer us- ing shifted windows. InProceedings of the IEEE/CVF international conference on computer vision. 10012–10022

    work page 2021

  27. [27]

    Martina Lofqvist and José Cano. 2020. Accelerating deep learning applications in space.arXiv preprint arXiv:2007.11089(2020)

    work page arXiv 2020

  28. [28]

    Giuseppe Masi, Davide Cozzolino, Luisa Verdoliva, and Giuseppe Scarpa. 2016. Pansharpening by convolutional neural networks.Remote Sensing8, 7 (2016), 594

    work page 2016

  29. [29]

    Abhas Maskey and Mengu Cho. 2020. CubeSatNet: Ultralight Convolutional Neural Network designed for on-orbit binary image classification on a 1U CubeSat. Engineering Applications of Artificial Intelligence96 (2020), 103952

    work page 2020

  30. [30]

    Qingyan Meng, Wenxu Shi, Sijia Li, and Linlin Zhang. 2023. PanDiff: A novel pansharpening method based on denoising diffusion probabilistic model.IEEE Transactions on Geoscience and Remote Sensing61 (2023), 1–17

    work page 2023

  31. [31]

    Xiangchao Meng, Nan Wang, Feng Shao, and Shutao Li. 2022. Vision transformer for pansharpening.IEEE Transactions on Geoscience and Remote Sensing60 (2022), 1–11

    work page 2022

  32. [32]

    Muhammed T Razzak, Gonzalo Mateo-Garcia, Gurvan Lecuyer, Luis Gómez- Chova, Yarin Gal, and Freddie Kalaitzis. 2023. Multi-spectral multi-image super- resolution of Sentinel-2 with radiometric consistency losses and its effect on building delineation.ISPRS Journal of Photogrammetry and Remote Sensing195 (2023), 1–13

    work page 2023

  33. [33]

    Francesco Salvetti, Vittorio Mazzia, Aleem Khaliq, and Marcello Chiaberge. 2020. Multi-image super resolution of remotely sensed images using residual attention deep neural networks.Remote Sensing12, 14 (2020), 2207

    work page 2020

  34. [34]

    Diego Valsesia and Enrico Magli. 2021. Permutation invariance and uncertainty in multitemporal image super-resolution.IEEE Transactions on Geoscience and Remote Sensing60 (2021), 1–12

    work page 2021

  35. [35]

    Ashish Vaswani, Noam Shazeer, Niki Parmar, Jakob Uszkoreit, Llion Jones, Aidan N Gomez, Łukasz Kaiser, and Illia Polosukhin. 2017. Attention is all you need.Advances in neural information processing systems30 (2017)

    work page 2017

  36. [36]

    Lucien Wald, Thierry Ranchin, and Marc Mangolini. 1997. Fusion of satellite images of different spatial resolutions: Assessing the quality of resulting images. Photogrammetric engineering and remote sensing63, 6 (1997), 691–699

    work page 1997

  37. [37]

    Shangguang Wang and Qing Li. 2023. Satellite computing: Vision and challenges. IEEE Internet of Things Journal10, 24 (2023), 22514–22529

    work page 2023

  38. [38]

    Shangguang Wang, Qing Li, Mengwei Xu, Xiao Ma, Ao Zhou, and Qibo Sun. 2021. Tiansuan constellation: An open research platform. In2021 IEEE International Conference on Edge Computing (EDGE). IEEE, 94–101

    work page 2021

  39. [39]

    Yancong Wei, Qiangqiang Yuan, Huanfeng Shen, and Liangpei Zhang. 2017. Boosting the accuracy of multispectral image pansharpening by learning a deep residual network.IEEE Geoscience and Remote Sensing Letters14, 10 (2017), 1795–1799

    work page 2017

  40. [40]

    Piper Wolters, Favyen Bastani, and Aniruddha Kembhavi. 2023. Zooming out on zooming in: Advancing super-resolution for remote sensing.arXiv preprint arXiv:2311.18082(2023)

    work page arXiv 2023

  41. [41]

    Qiang Wu, Hongfei Zeng, Jin Zhang, Weishi Li, and Haojie Xia. 2024. A hybrid network of CNN and transformer for subpixel shifting-based multi-image super- resolution.Optics and Lasers in Engineering182 (2024), 108458

    work page 2024

  42. [42]

    Yinghui Xing, Litao Qu, Shizhou Zhang, Kai Zhang, Yanning Zhang, and Lorenzo Bruzzone. 2024. CrossDiff: Exploring self-supervised representation of pansharp- ening via cross-predictive diffusion model.IEEE Transactions on Image Processing 33 (2024), 5496–5509

    work page 2024

  43. [43]

    Chen Yang, Qibo Sun, Qiyang Zhang, Hao Lu, Claudio A Ardagna, Shangguang Wang, and Mengwei Xu. 2024. Towards Efficient Satellite Computing Through Adaptive Compression.IEEE Transactions on Services Computing17, 6 (2024), 4411–4422

    work page 2024

  44. [44]

    Junfeng Yang, Xueyang Fu, Yuwen Hu, Yue Huang, Xinghao Ding, and John Paisley. 2017. PanNet: A deep network architecture for pan-sharpening. In Proceedings of the IEEE international conference on computer vision. 5449–5457

    work page 2017

  45. [45]

    Qiangqiang Yuan, Yancong Wei, Xiangchao Meng, Huanfeng Shen, and Liangpei Zhang. 2018. A multiscale and multidepth convolutional neural network for remote sensing imagery pan-sharpening.IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing11, 3 (2018), 978–989

    work page 2018

  46. [46]

    Roberta H Yuhas, Alexander FH Goetz, and Joe W Boardman. 1992. Discrimina- tion among semi-arid landscape endmembers using the spectral angle mapper (SAM) algorithm. InJPL, Summaries of the Third Annual JPL Airborne Geoscience Workshop. Volume 1: A VIRIS Workshop

    work page 1992

  47. [47]

    Bing Zhang, Yuanfeng Wu, Boya Zhao, Jocelyn Chanussot, Danfeng Hong, Jing Yao, and Lianru Gao. 2022. Progress and challenges in intelligent remote sensing satellite systems.IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing15 (2022), 1814–1822

    work page 2022

  48. [48]

    Yulun Zhang, Kunpeng Li, Kai Li, Lichen Wang, Bineng Zhong, and Yun Fu. 2018. Image super-resolution using very deep residual channel attention networks. In Proceedings of the European conference on computer vision (ECCV). 286–301

    work page 2018

  49. [49]

    Yuxuan Zheng, Jiaojiao Li, Yunsong Li, Jie Guo, Xianyun Wu, and Jocelyn Chanus- sot. 2020. Hyperspectral pansharpening using deep prior and dual attention residual network.IEEE transactions on geoscience and remote sensing58, 11 (2020), 8059–8076

    work page 2020

  50. [50]

    Man Zhou, Jie Huang, Yanchi Fang, Xueyang Fu, and Aiping Liu. 2022. Pan- sharpening with customized transformer and invertible neural network. InPro- ceedings of the AAAI conference on artificial intelligence, Vol. 36. 3553–3561

    work page 2022

  51. [51]

    Chunyu Zhu, Rongyuan Dai, Liwei Gong, Liangbo Gao, Na Ta, and Qiong Wu

    work page

  52. [52]

    An adaptive multi-perceptual implicit sampling for hyperspectral and multispectral remote sensing image fusion.International Journal of Applied Earth Observation and Geoinformation125 (2023), 103560

    work page 2023

  53. [53]

    Zhe Zhu and Curtis E Woodcock. 2012. Object-based cloud and cloud shadow detection in Landsat imagery.Remote sensing of environment118 (2012), 83–94

    work page 2012

  54. [54]

    Maciej Ziaja, Pawel Kowaleczko, Tomasz Tarasiewicz, Daniel Kostrzewa, and Michal Kawulok. 2025. Hyperspectral Pansharpening Enhanced With Multi- Image Super-Resolution for PRISMA Data.IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing(2025). A Hyperparameter Experiments We addressRQ3by varying two key hyperparameters. We firs...

    work page arXiv 2025