Pith. sign in

REVIEW 3 major objections 7 minor 58 references

Cloud Removal With PolSAR-Optical Data Fusion Using A Two-Flow Residual Network

T0 review · 3 major / 7 minor · reviewed 2026-08-10 · deepseek-v4-flash

Challenge this review Re-run · record.json Download PDF Read on arXiv ↗ Review JSON

Pith's one-line read This paper claims that a two-flow residual network fusing full-polarimetric SAR features with cloudy optical images reconstructs cloud-free optical imagery more accurately than existing methods, reporting state-of-the-art PSNR, SSIM…

desk verdict The PFSAR-only variant is the actual method despite the BCFSAR+PFSAR framing; the main barrier is the evaluation protocol, not the architecture. read the letter →

arxiv 2501.07901 v1 pith:GQHLLDHD submitted 2025-01-14 cs.CV eess.IV

Yuxi Wang , Wenjuan Zhang , Bing Zhang This is my paper · ORCID
classification cs.CVeess.IV MSC 68T0768U1094A08
keywords cloudremovalPolSARoptical-SARfusiontwo-flowresidualnetworkdynamicfiltersattentionmechanismimagerestorationpolarimetricfeatures
verification ladder T0 review T1 audit T2 compute T3 formal

The pith

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

The reading

This paper attempts to establish that full-polarimetric SAR (PolSAR) data can serve as a stronger auxiliary source than single- or dual-polarization SAR for reconstructing optical satellite images hidden by clouds. To do this, it introduces PODF-CR, a two-flow residual network that encodes PolSAR and optical images in parallel, suppresses speckle with decoupled dynamic filters, and fuses the two streams with cross-skip fusion blocks and an attention-based refinement stage. On the OPT-BCFSAR-PFSAR airborne dataset, the method reports the best quantitative results among seven methods, with PSNR 34.992, SSIM 0.892, correlation coefficient 0.922, and SAM 0.916, and it retains an advantage at heavy cloud coverage up to 80-100%. A sympathetic reader would care because the paper shows that polarization and scattering features extracted from the coherence matrix can carry enough ground information to restore fine texture details that optical-only or single-polarization SAR methods miss.

What carries the argument

The load-bearing mechanism is the two-flow fusion of PolSAR-derived features with optical features. The PolSAR branch is built around SCDF (coupled spatial and channel dynamic filters), which generates per-pixel spatial filters and per-channel filters from the input to suppress speckle before features propagate; MMCF (multi-modality cross fusion) blocks use cross-modality dense skip connections to let each branch see the other's feature maps at multiple encoder scales; and MMRF (multi-modality refinement fusion) blocks apply spatial-channel attention plus a cross-modal weighted refinement unit that computes pixel-wise correlations between optical, PolSAR, and fusion features to refine the fused representation globally. This combination is what lets the network use polarization and scattering information to fill cloud gaps while preserving cloud-free regions.

What would settle it

Take the trained PODF-CR model and test it on real cloud-contaminated optical images from the same area where the cloud mask is not known and coregistration errors exceed the stated 2-pixel tolerance; if PSNR and SSIM against a later cloud-free acquisition drop to the level of the DSen2-CR or USSRN-CR baselines, then the reported state-of-the-art numbers depend on the simulated-mask training setup rather than on the fusion architecture itself.

Watch

Extended reading notes

Core claim

The central claim is that the proposed PODF-CR network reconstructs cloud-free optical images from four aligned inputs: the cloudy optical image, a 3-band backscatter coefficient image (BCFSAR), a 9-band polarization feature image (PFSAR), and a cloud mask used in the loss. The architecture is an encoder-decoder with two parallel branches; the optical branch uses gated convolutions to avoid propagating cloud pixels, the PolSAR branch uses coupled spatial-channel dynamic filters to suppress speckle, and multi-modality cross fusion (MMCF) blocks pass each modality's features into the other branch at several scales via cross-skip connections. The resulting fusion features are refined by a multi-modality refinement fusion (MMRF) block with spatial-channel attention and a cross-modal weighted refinement unit, and the decoder applies atrous spatial pyramid pooling to combine multi-scale context before output. The loss combines global L1, a cloud-masked local L1, and SSIM. The paper reports that this configuration outperforms six comparison algorithms on the dataset, and ablation experiments attribute the gains to the polarization features, the dynamic filters, the gated convolutions, the fusion modules, and the multi-scale pooling.

Load-bearing premise

The load-bearing premise is that randomly generated cloud masks over a single-date airborne scene faithfully mimic real cloud corruption, with optical and PolSAR images registered within about 2 pixels, so the measured reconstruction gains transfer to genuine cloud-covered imagery.

Editorial extensions

If this is right

  • Full-polarimetric SAR input, represented through backscatter coefficients and polarization features from the coherence matrix, becomes an effective auxiliary modality for optical cloud removal.
  • The reported architecture is a concrete recipe: parallel optical and PolSAR encoders, gated convolutions on the optical side, dynamic filters on the PolSAR side, cross-skip fusion blocks, attention-based refinement, and multi-scale decoding.
  • Ablation results indicate that polarization feature images (PFSAR) carry more restoration value than backscatter coefficient images (BCFSAR), with PFSAR-only reaching the same top PSNR of 34.992 as the full model.
  • Performance degrades gracefully with increasing cloud coverage, with the method still reconstructing building and traffic-route structure at 80-100% cloud cover where other models blur.
  • Removing the PolSAR auxiliary data entirely drops PSNR by about 2.2 dB, supporting the claim that the PolSAR branch is the source of the missing-area information.

Reading between the lines

Editorial extensions of the paper, not claims the author makes directly.

  • If the PFSAR-only result generalizes, a testable extension is that BCFSAR backscatter images may be redundant or even detrimental for this setting; future datasets could treat them as an auxiliary denoising channel rather than primary input.
  • The method's reliance on a coherence-matrix representation suggests that other polarimetric decompositions (for example, Cloude-Pottier, Freeman-Durden, or Yamaguchi components) could be substituted for the current nine-band PFSAR input and compared directly.
  • Because the training data is a single airborne scene with simulated random cloud masks, real-world deployment would require validation on multi-temporal, satellite-acquired cloudy images; the reported 2-pixel registration tolerance between optical and PolSAR is a likely practical constraint.
  • The 58.86 million parameters and roughly 419 GFLOPs per 256x256 tile imply that satellite-scale processing would need tiling or a lighter variant, a direction the paper does not address.
Share X Bluesky LinkedIn Reddit HN

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, and a circularity audit.

Referee Report

3 major / 7 minor

Summary. The manuscript proposes PODF-CR, a two-flow encoder-decoder network for cloud removal that fuses cloudy optical images with PolSAR-derived backscatter coefficient images (BCFSAR) and polarization feature images (PFSAR). The PolSAR branch uses spatial and channel dynamic filters for denoising, and the network introduces multi-modality cross fusion (MMCF), multi-modality refinement fusion (MMRF), and atrous spatial pyramid pooling (ASPP). The method is evaluated on a new airborne OPT-BCFSAR-PFSAR dataset of 2030 image pairs with simulated cloud masks, and the paper claims state-of-the-art results in PSNR, SSIM, CC, and SAM relative to six baselines.

Significance. If the reported numbers are correct and reproducible, the paper would offer a useful architecture and a new PolSAR-feature dataset for SAR-optical cloud removal; the dynamic-filter denoising and cross-modal fusion components are sensible, and the per-category and per-cloud-cover analysis is more detailed than is common in this literature. However, the central quantitative claim is currently undermined by an internal inconsistency between Tables 3 and 4, and the evaluation protocol does not separate model selection from final testing. As a result, the significance of the contribution cannot be assessed from the manuscript as written. No code or dataset repository is indicated, which further limits immediate verification.

major comments (3)
  1. [§4.2.1 / Table 3 and §4.3.1.1 / Table 4] The headline result in Table 3 (PODF-CR: PSNR 34.992, SSIM 0.892, CC 0.922, SAM 0.916) is numerically identical to the 'w PFSAR' ablation row in Table 4, while the full model with both BCFSAR and PFSAR ('w PolSAR') is reported as 34.874/0.890/0.921/0.922. Section 4.3.1.1 further states that adding BCFSAR feature images may introduce noise to residual learning and decrease reconstruction accuracy. The main comparison table therefore appears to attribute the PFSAR-only variant to the full two-flow PODF-CR architecture described in Section 3.1. This must be reconciled: either Table 3 is mislabeled, or the two tables were produced under different experimental conditions. In either case, the current text does not support the claim that the proposed two-flow BCFSAR+PFSAR fusion produces the advertised state-of-the-art numbers.
  2. [§4.1.3 and §4.3.2] Hyperparameters are selected on the test split: the learning rate is found by 'systematic search', and lambda1 and lambda2 are chosen in Fig. 13 by maximizing PSNR, with no separate validation set described anywhere in the 80/20 train/test split. This makes the reported quantitative comparisons and final model selection optimistic and makes the protocol impossible to reproduce as a fixed evaluation. Please introduce a validation split or nested cross-validation, select hyperparameters and input configurations using validation data only, and report test-set metrics exactly once after fixing those choices.
  3. [§4.1.1] The dataset consists of a single airborne scene acquired on one date, with clouds simulated by randomly generated masks; no real cloudy optical images or independent acquisition dates are used. Since the reconstruction loss is evaluated only on these simulated masks, the claim that the method outperforms existing methods for actual cloud removal is not yet supported. I recommend evaluating on a public benchmark such as SEN12MS-CR or including real cloudy optical images with co-registered PolSAR data, and also reporting sensitivity to registration errors beyond the stated 2-pixel tolerance.
minor comments (7)
  1. [Table 3] The CC column is marked with a down arrow, but Section 4.1.2 states that higher CC values indicate better performance; the arrow should be an up arrow.
  2. [§3.2.4 and Figure 4] The attention unit is called SCRU in the text and SCAU in Figure 4, and the fusion block is called MMRF in the text but MRFB in Figure 4(a); please unify the notation.
  3. [§4.3.3] The text says the model has a 'parameter number of 58.86' without specifying the unit; it should state 58.86M, and the FLOPs figures should state the exact input size used for measurement.
  4. [§3.2.5] The ASPP paragraph contains a bare '(?)' placeholder where a citation is expected; please insert the appropriate reference.
  5. [§1.1 and throughout] The dataset name is given as OPT-FPBCSAR-PFSAR in the contribution list and OPT-BCFSAR-PFSAR elsewhere; please make the name consistent.
  6. [§2.1] The reference list appears to contain the same Cheng et al. paper twice as 2013a and 2013b; please verify and merge duplicate entries.
  7. [Figure 13] The axis labels in Figure 13 contain stray characters and are difficult to read; please regenerate the figure with clean mathematical notation.

Circularity Check

1 steps flagged · score 3.0 of 10

No algebraic circularity; mild evaluation-protocol circularity from selecting hyperparameters and the final input configuration on the same test split used for the headline metrics.

  1. fitted input called prediction [Section 4.1.3 (Implementation Details) and Section 4.3.2 (Parameter Sensitive Analysis), with the train/test split defined in Section 4.1.1]
    "After systematic search, the optimal learning rate was found to be 7 × 10−5. ... The empirical weights are set to λ1 = 10 and λ2 = 1. ... We mainly analyzed the weights λ1 and λ2 of the local loss and structural loss in the loss function in order to fine-tune the model to the optimal state. For quantitative evaluation, we compared models with different weights using PSNR."

    The paper defines only a training split (80%) and a test split (20%), with no validation split described in Section 4.1.1. The 'systematic search' for the learning rate and the PSNR-based choice of λ1=10 and λ2=1 therefore use the same test split whose PSNR is reported as the headline performance in Table 3. The reported PSNR is the objective used to select the hyperparameters, so the advertised numbers are not an independent out-of-sample prediction. Section 4.3.1.1 similarly selects the PFSAR-only input variant by comparing PSNR on this same test split, and that variant's row becomes the PODF-CR result in Table 3. This is a mild circularity of evaluation protocol rather than an algebraic derivation collapse.

full rationale

The paper contains no equation whose output is defined in terms of the claimed result, and no load-bearing self-citation: the cited USSRN-CR (Wang et al., 2023) is used as a comparison baseline, and the architectural components (SCDF, MMCF, MMRF, ASPP) are standard learned modules evaluated by ablation. The central empirical comparison is self-contained relative to the cited literature. The only circularity-like feature is the evaluation protocol: hyperparameters and the final input configuration are selected by PSNR on the same 20% test split that generates the headline results, making the reported gains optimistic. This is a methodological weakness and a mild form of fitting the evaluation metric, but it does not reduce the derivation to its inputs. The paper itself notes as a limitation only the need for multi-temporal data in future work, not this test-set selection issue.

Assumptions & free parameters 7 free parameters · 6 assumptions · 0 invented entities

The network is a supervised deep learning system; its central performance claim rests on the data generation and evaluation protocol. The ledger lists the fitted loss weights and optimizer settings that were tuned on the test split, the domain assumptions about registration and random cloud masks, and records that no new physical entities are proposed. The introduced components (SCDF, MMCF, MMRF) are neural network modules, not postulated entities, and the OPT-BCFSAR-PFSAR dataset is a data artifact, not an invented entity.

free parameters (7)
  • lambda1 (local loss weight) = 10
    Tuned by maximum PSNR in Section 4.3.2; no separate validation set.
  • lambda2 (SSIM loss weight) = 1
    Selected by PSNR sweep in Section 4.3.2 after fixing lambda1.
  • Learning rate = 7e-5
    Chosen by systematic search in Section 4.1.3.
  • Residual scaling constant = 0.1
    Used in RB-GC to stabilize training; taken from Meraner et al. 2020, not data-fitted.
  • Encoder channels (64,128,256) = 64/128/256
    Architecture capacity choice in Section 3.1; no ablation for channel width.
  • ASPP dilation rates = 6, 12, 18
    Hand-selected standard ASPP configuration from DeepLabv3+; no ablation on dilation rates.
  • Adam beta1, beta2 = 0.5, 0.999
    Fixed optimizer hyperparameters in Section 4.1.3; not fitted but affect training.
assumptions (6)
  • domain assumption Registered optical and PolSAR images have pixel-level correspondence (registration deviation less than 2 pixels).
    Invoked in Section 4.1.1 to justify using PolSAR to reconstruct optical pixels; misregistration would break the supervision and the fusion.
  • ad hoc to paper Randomly generated cloud masks over the same scene faithfully simulate real cloud corruption.
    Section 4.1.1 creates cloudy images by overlaying random masks; no real cloudy/cloud-free pairs are used, so transferability to actual clouds is assumed.
  • domain assumption The 9-band PFSAR and 3-band BCFSAR features derived from the coherence matrix retain sufficient scattering and polarization information for optical reconstruction.
    Section 3.1 and Section 4.3.1.1 rely on these hand-crafted feature images as the auxiliary modality; no comparison against using raw complex PolSAR data.
  • domain assumption The airborne single-date Zhangye scene statistics are representative of cloud-removal conditions elsewhere.
    The entire dataset is from one area and one acquisition date (Section 4.1.1); generalization to other sensors and regions is assumed.
  • domain assumption HV and VH polarization channels are identical, so only HV is used.
    Section 4.1.1 drops VH on reciprocity; standard in SAR but still a modeling choice.
  • domain assumption L1, masked L1, and SSIM losses are sufficient proxies for perceptual reconstruction quality.
    Section 3.2.6 defines the training objective; no human study or task-based evaluation is provided.

how reviews work

0 comments
Cite this review

Pith. "Pith review of Cloud Removal With PolSAR-Optical Data Fusion Using A Two-Flow Residual Network." pith.science (2026). https://pith.science/paper/GQHLLDHD

@misc{pith2026250107901,
  author       = {Pith},
  title        = {Pith review of: Cloud Removal With PolSAR-Optical Data Fusion Using A Two-Flow Residual Network},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/GQHLLDHD}},
  note         = {Machine review of arXiv:2501.07901}
}
read the original abstract

Optical remote sensing images play a crucial role in the observation of the Earth's surface. However, obtaining complete optical remote sensing images is challenging due to cloud cover. Reconstructing cloud-free optical images has become a major task in recent years. This paper presents a two-flow Polarimetric Synthetic Aperture Radar (PolSAR)-Optical data fusion cloud removal algorithm (PODF-CR), which achieves the reconstruction of missing optical images. PODF-CR consists of an encoding module and a decoding module. The encoding module includes two parallel branches that extract PolSAR image features and optical image features. To address speckle noise in PolSAR images, we introduce dynamic filters in the PolSAR branch for image denoising. To better facilitate the fusion between multimodal optical images and PolSAR images, we propose fusion blocks based on cross-skip connections to enable interaction of multimodal data information. The obtained fusion features are refined through an attention mechanism to provide better conditions for the subsequent decoding of the fused images. In the decoding module, multi-scale convolution is introduced to obtain multi-scale information. Additionally, to better utilize comprehensive scattering information and polarization characteristics to assist in the restoration of optical images, we use a dataset for cloud restoration called OPT-BCFSAR-PFSAR, which includes backscatter coefficient feature images and polarization feature images obtained from PoLSAR data and optical images. Experimental results demonstrate that this method outperforms existing methods in both qualitative and quantitative evaluations.

Figures

Figures reproduced from arXiv: 2501.07901 by the authors.

Figure 1
Figure 1. Overview of the proposed PolSAR-Optical data fusion based cloud removal (PODF-CR) algorithm. In the decoder module, the fusion features extracted from PolSAR images and cloudy optical images are decoded to reconstruct the contaminated areas. An atrous spatial pyramid pooling (ASPP) module, two upsampling modules, four residual blocks, and a convolutional layer are contained. The upsampling module also includes a con… view at source ↗
Figure 2
Figure 2. Illustration of the residual blocks based on gated convolutions (RB-GC), the residual blocks based on dynamic filters (RB-DF) and the spatial and channel dynamic filters (SCDF). (a) RB-GC. (b) RB-DF. (c) SCDF [PITH_FULL_IMAGE:figures/full_fig_p008_2.png] view at source ↗
Figure 3
Figure 3. Illustration of the multi-modality cross fusion block(MMCF). 𝐹 𝑂𝑝𝑡 2 = 𝐺𝑐𝑜𝑛𝑣(𝐹 𝑂𝑝𝑡 1 + (𝑊 𝑃 𝑜𝑙𝑆𝐴𝑅 12 𝐹 𝑃 𝑜𝑙𝑆𝐴𝑅 1 + 𝑏 𝑃 𝑜𝑙𝑆𝐴𝑅 12 )) (6) Where 𝑊 𝑂𝑝𝑡 𝑖𝑗 is the weight from the 𝑖 𝑡ℎ optical block to the 𝑗 𝑡ℎ PolSAR feature block, and 𝑏 𝑂𝑝𝑡 𝑖𝑗 is the corresponding bias. 𝑊 𝑃 𝑜𝑙𝑆𝐴𝑅 𝑖𝑗 is the weight from the 𝑗 𝑡ℎ optical block to the 𝑖 𝑡ℎ PolSAR feature block. Similar to the first layer skip fusion, the output for the secon… view at source ↗
Figures from the paper (10 more)
Figure 4
Figure 4. Figure 4: Illustration of the multi-modality refinement fusion. (a) Multi-modal refinement fusion block (MRFB). (b) Spatial-channel attention unit(SCAU). global context information. The dilation rates of the atrous convolutions are 6, 12, and 18, respectively. Subsequently, the …
Figure 5
Figure 5. Figure 5: Illustration of the atrous spatial pyramid pooling (ASPP) structure. Y. Wang, et al. Page 10 of 21 [PITH_FULL_IMAGE:figures/full_fig_p010_5.png]
Figure 6
Figure 6. Figure 6: Example of the main categories of the OPT-BCFSAR-PFSAR dataset. model uses a cGAN network to generate cloud-free optical data from cloud-damaged optical input and auxiliary SAR images. The DSen2-CR model and the USSRN-CR model take concatenated PolSAR data and degraded…
Figure 7
Figure 7. Figure 7: Experimental results of the PODF-CR model to state-of-the-art methods in dataset. (a) Cloudy images. (b) PFSAR images. (c) Results of the SpAGAN model. (d) Results of the pix2pix model. (e) Results of the SAR-opt-cGAN model. (f) Results of the DSen2-CR model. (g) Resul…
Figure 8
Figure 8. Figure 8: Experimental results of the PODF-CR model to state-of-the-art methods based on different category in dataset. (a) Cloudy images. (b) PFSAR images. (c) Results of the SpAGAN model. (d) Results of the pix2pix model. (e) Results of the SAR-opt-cGAN model. (f) Results of t…
Figure 9
Figure 9. Figure 9: Experimental results of the PODF-CR model to state-of-the-art methods based on different cloud coverage in dataset. (a) Cloudy images. (b) PFSAR images. (c) Results of the SpAGAN model. (d) Results of the pix2pix model. (e) Results of the SAR-opt-cGAN model. (f) Result…
Figure 10
Figure 10. Figure 10: Comparisons of the PODF-CR model to state-of-the-art methods on different category and cloud coverage in terms of the PSNR, SSIM, SAM, and MAE quality metrics [PITH_FULL_IMAGE:figures/full_fig_p017_10.png]
Figure 11
Figure 11. Figure 11: Ablation experiments based on input data of PODF-CR algorithm. (a) Cloudy images. (b) BCFSAR images. (c) PFSAR images. (d) Results of the w/o PolSAR model. (e) Results of the w PolSAR model. (f) Results of the w BCFSAR model. (g) Results of the w PFSAR model. (h) Grou…
Figure 12
Figure 12. Figure 12: Ablation experiments based on the module of PODF-CR algorithm. (a) Cloudy images. (b) PFSAR images. (c) Results of the w/o SCDF model. (d) Results of the w/o GC model. (e) Results of the w/o MMCF model. (f) Results of the w/o MMRF model. (g) Results of the w/o ASPP mo…
Figure 13
Figure 13. Figure 13: Results of the proposed model with different weight. (a) 𝜆1. (b) 𝜆2 [PITH_FULL_IMAGE:figures/full_fig_p019_13.png]

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

58 extracted references · 55 canonical work pages

  1. [1]

    write newline

    " write newline "" before.all 'output.state := FUNCTION list.label.output make.list.label ", " * write FUNCTION article output.bibitem list.label.output list.year output.nonnull new.block author "author" item.check title.field field.used = skip format.title "title" output.check if new.block crossref missing journal "journal" duplicate item.check pages emp...

    work page

  2. [2]

    and Liao, H., 2019

    Anastasopoulos, A., Kumar, S. and Liao, H., 2019. Neural language modeling with visual features. arXiv preprint arXiv:1903.02930

    arXiv 2019

  3. [3]

    Ballester-Berman, J. D. and Lopez-Sanchez, J. M., 2009. Applying the freeman--durden decomposition concept to polarimetric sar interferometry. IEEE Transactions on Geoscience and Remote Sensing 48(1), pp. 466--479

    work page 2009

  4. [4]

    D., Happ, P

    Bermudez, J. D., Happ, P. N., Oliveira, D. A. B. and Feitosa, R. Q., 2018. Sar to optical image synthesis for cloud removal with generative adversarial networks. ISPRS Annals of the Photogrammetry, Remote Sensing and Spatial Information Sciences 4, pp. 5--11

    work page 2018

  5. [5]

    and Philips, W., 2006

    Borghys, D., Yvinec, Y., Perneel, C., Pizurica, A. and Philips, W., 2006. Supervised feature-based classification of multi-channel sar images. Pattern Recognition Letters 27(4), pp. 252--258

    work page 2006

  6. [6]

    E., Gao, F

    Chen, J., Zhu, X., Vogelmann, J. E., Gao, F. and Jin, S., 2011. A simple and effective method for filling gaps in landsat etm+ slc-off images. Remote sensing of environment 115(4), pp. 1053--1064

    work page 2011

  7. [7]

    and Adam, H., 2018

    Chen, L.-C., Zhu, Y., Papandreou, G., Schroff, F. and Adam, H., 2018. Encoder-decoder with atrous separable convolution for semantic image segmentation. In: Proceedings of the European conference on computer vision (ECCV), pp. 801--818

    work page 2018

  8. [8]

    and Li, P., 2013a

    Cheng, Q., Shen, H., Zhang, L. and Li, P., 2013a. Inpainting for remotely sensed images with a multichannel nonlocal total variation model. IEEE Transactions on Geoscience and Remote Sensing 52(1), pp. 175--187

    work page

Show all 58 references
  1. [9]

    and Li, P., 2013b

    Cheng, Q., Shen, H., Zhang, L. and Li, P., 2013b. Inpainting for remotely sensed images with a multichannel nonlocal total variation model. IEEE Transactions on Geoscience and Remote Sensing 52(1), pp. 175--187

  2. [10]

    and Fan, K.-C., 2018

    Chiang, C.-Y., Chen, K.-S., Chu, C.-Y., Chang, Y.-L. and Fan, K.-C., 2018. Color enhancement for four-component decomposed polarimetric sar image based on a cie-lab encoding. Remote Sensing 10(4), pp. 545

    2018

  3. [11]

    Cloude, S. R. and Pottier, E., 1997. An entropy based classification scheme for land applications of polarimetric sar. IEEE transactions on geoscience and remote sensing 35(1), pp. 68--78

    1997

  4. [12]

    and Toyama, K., 2004

    Criminisi, A., P \'e rez, P. and Toyama, K., 2004. Region filling and object removal by exemplar-based image inpainting. IEEE Transactions on image processing 13(9), pp. 1200--1212

    2004

  5. [13]

    and Schmullius, C., 2013

    Eckardt, R., Berger, C., Thiel, C. and Schmullius, C., 2013. Removal of optically thick clouds from multi-spectral satellite images using multi-frequency sar data. Remote Sensing 5(6), pp. 2973--3006

    2013

  6. [14]

    and Erik, L., 2020

    Erith, M., Alfonso, Z. and Erik, L., 2020. A multi-sensor approach to separate palm oil plantations from forest cover using ndfi and a modified pauli decomposition technique. In: IGARSS 2020-2020 IEEE International Geoscience and Remote Sensing Symposium, IEEE, pp. 4481--4484

    2020

  7. [15]

    and Wang, P., 2023

    Fang, Z., Zhang, G., Dai, Q., Xue, B. and Wang, P., 2023. Hybrid attention-based encoder--decoder fully convolutional network for polsar image classification. Remote Sensing 15(2), pp. 526

    2023

  8. [16]

    and Schmitt, M., 2019

    Fuentes Reyes, M., Auer, S., Merkle, N., Henry, C. and Schmitt, M., 2019. Sar-to-optical image translation based on conditional generative adversarial networks—optimization, opportunities and limits. Remote Sensing 11(17), pp. 2067

    2019

  9. [17]

    and Hirosawa, H., 2001

    Fukuda, S. and Hirosawa, H., 2001. Support vector machine classification of land cover: Application to polarimetric sar data. In: IGARSS 2001. Scanning the Present and Resolving the Future. Proceedings. IEEE 2001 International Geoscience and Remote Sensing Symposium (Cat. No. ...

    2001

  10. [18]

    and Su, X., 2020

    Gao, J., Yuan, Q., Li, J., Zhang, H. and Su, X., 2020. Cloud removal with fusion of high resolution optical and sar images using generative adversarial networks. Remote Sensing 12(1), pp. 191

    2020

  11. [19]

    and Zhu, X., 2018

    Grohnfeldt, C., Schmitt, M. and Zhu, X., 2018. A conditional generative adversarial network to fuse sar and multispectral optical data for cloud removal from sentinel-2 images. In: IGARSS 2018-2018 IEEE International Geoscience and Remote Sensing Symposium, IEEE, pp. 1726--1729

    2018

  12. [20]

    and Cloude, S

    Hajnsek, I., Pottier, E. and Cloude, S. R., 2003. Inversion of surface parameters from polarimetric sar. IEEE Transactions on Geoscience and Remote Sensing 41(4), pp. 727--744

    2003

  13. [21]

    and Zhang, S., 2023

    Han, S., Wang, J. and Zhang, S., 2023. Former-cr: A transformer-based thick cloud removal method with optical and sar imagery. Remote Sensing 15(5), pp. 1196

    2023

  14. [22]

    and Sun, J., 2014

    He, K. and Sun, J., 2014. Image completion approaches using the statistics of similar patches. IEEE transactions on pattern analysis and machine intelligence 36(12), pp. 2423--2435

    2014

  15. [23]

    and Zhang, B., 2020

    Hong, D., Gao, L., Yokoya, N., Yao, J., Chanussot, J., Du, Q. and Zhang, B., 2020. More diverse means better: Multimodal deep learning meets remote-sensing imagery classification. IEEE Transactions on Geoscience and Remote Sensing 59(5), pp. 4340--4354

    2020

  16. [24]

    Batch normalization: Accelerating deep network training by reducing internal covariate shift

    Ioffe, S., 2015. Batch normalization: Accelerating deep network training by reducing internal covariate shift. arXiv preprint arXiv:1502.03167

    2015 arXiv

  17. [25]

    and Efros, A

    Isola, P., Zhu, J.-Y., Zhou, T. and Efros, A. A., 2017. Image-to-image translation with conditional adversarial networks. In: Proceedings of the IEEE conference on computer vision and pattern recognition, pp. 1125--1134

    2017

  18. [26]

    and Gool, L

    Jia, X., De Brabandere, B., Tuytelaars, T. and Gool, L. V., 2016. Dynamic filter networks. Advances in neural information processing systems

    2016

  19. [27]

    D., Platnick, S., Menzel, W

    King, M. D., Platnick, S., Menzel, W. P., Ackerman, S. A. and Hubanks, P. A., 2013. Spatial and temporal distribution of clouds observed by modis onboard the terra and aqua satellites. IEEE transactions on geoscience and remote sensing 51(7), pp. 3826--3852

    2013

  20. [28]

    and Ba, J., 2014

    Kingma, D. and Ba, J., 2014. Adam: A method for stochastic optimization. Computer Science

    2014

  21. [29]

    and Kussul, O., 2014

    Kussul, N., Skakun, S., Shelestov, A. and Kussul, O., 2014. The use of satellite sar imagery to crop classification in ukraine within jecam project. In: 2014 IEEE Geoscience and Remote Sensing Symposium, IEEE, pp. 1497--1500

    2014

  22. [30]

    and Yamagata, Y., 2020

    La, Y., Bagan, H. and Yamagata, Y., 2020. Urban land cover mapping under the local climate zone scheme using sentinel-2 and palsar-2 data. Urban Climate 33, pp. 100661

    2020

  23. [31]

    and Pottier, E., 2017

    Lee, J.-S. and Pottier, E., 2017. Polarimetric radar imaging: from basics to applications. CRC press

    2017

  24. [32]

    and Fang, L., 2019

    Li, X., Wang, L., Cheng, Q., Wu, P., Gan, W. and Fang, L., 2019. Cloud removal in remote sensing images using nonnegative matrix factorization and error correction. ISPRS journal of photogrammetry and remote sensing 148, pp. 103--113

    2019

  25. [33]

    and Lei, B., 2018

    Liu, L. and Lei, B., 2018. Can sar images and optical images transfer with each other? In: IGARSS 2018-2018 IEEE International Geoscience and Remote Sensing Symposium, IEEE, pp. 7019--7022

    2018

  26. [34]

    and Lin, L., 2021

    Liu, L., Chen, J., Wu, H., Li, G., Li, C. and Lin, L., 2021. Cross-modal collaborative representation learning and a large-scale rgbt benchmark for crowd counting. In: Proceedings of the IEEE/CVF conference on computer vision and pattern recognition, pp. 4823--4833

    2021

  27. [35]

    and Pani, S., 2015

    Maity, A., Pattanaik, A., Sagnika, S. and Pani, S., 2015. A comparative study on approaches to speckle noise reduction in images. In: 2015 International Conference on Computational Intelligence and Networks, IEEE, pp. 148--155

    2015

  28. [36]

    Meraner, A., Ebel, P., Zhu, X. X. and Schmitt, M., 2020. Cloud removal in sentinel-2 imagery using a deep residual neural network and sar-optical data fusion. ISPRS Journal of Photogrammetry and Remote Sensing 166, pp. 333--346

    2020

  29. [37]

    T., Chen, J., Sharlet, D., Ng, R

    Mildenhall, B., Barron, J. T., Chen, J., Sharlet, D., Ng, R. and Carroll, R., 2018. Burst denoising with kernel prediction networks. In: Proceedings of the IEEE conference on computer vision and pattern recognition, pp. 2502--2510

    2018

  30. [38]

    and Papathanassiou, K

    Moreira, A., Prats-Iraola, P., Younis, M., Krieger, G., Hajnsek, I. and Papathanassiou, K. P., 2013. A tutorial on synthetic aperture radar. IEEE Geoscience and remote sensing magazine 1(1), pp. 6--43

    2013

  31. [39]

    and Liu, F., 2017

    Niklaus, S., Mai, L. and Liu, F., 2017. Video frame interpolation via adaptive convolution. In: Proceedings of the IEEE conference on computer vision and pattern recognition, pp. 670--679

    2017

  32. [40]

    E., Olayinka, D

    Okwuashi, O., Ndehedehe, C. E., Olayinka, D. N., Eyoh, A. and Attai, H., 2021. Deep support vector machine for polsar image classification. International Journal of Remote Sensing 42(17), pp. 6498--6536

    2021

  33. [41]

    Cloud removal for remote sensing imagery via spatial attention generative adversarial network

    Pan, H., 2020. Cloud removal for remote sensing imagery via spatial attention generative adversarial network. arXiv preprint arXiv:2009.13015

    2020 arXiv

  34. [42]

    G., Chen, H., Moa, B., Hobart, G

    Richardson, A., Goodenough, D. G., Chen, H., Moa, B., Hobart, G. and Myrvold, W., 2010. Unsupervised nonparametric classification of polarimetric sar data using the k-nearest neighbor graph. In: 2010 IEEE International Geoscience and Remote Sensing Symposium, IEEE, pp. 1867--1870

    2010

  35. [43]

    and Haiying, L., 2016

    Ruliang, Y., Bowei, D. and Haiying, L., 2016. Polarization hierarchy and system operating architecture for polarimetric synthetic aperture radar. Journal of radars 5(2), pp. 132--142

    2016

  36. [44]

    and Zhu, X

    Schmitt, M., Tupin, F. and Zhu, X. X., 2017. Fusion of sar and optical remote sensing data—challenges and recent trends. In: 2017 IEEE International Geoscience and Remote Sensing Symposium (IGARSS), IEEE, pp. 5458--5461

    2017

  37. [45]

    Partial differential equation methods for image inpainting

    Sch \"o nlieb, C.-B., 2015. Partial differential equation methods for image inpainting. Vol. 29, Cambridge University Press

    2015

  38. [46]

    and Chan, T

    Shen, J. and Chan, T. F., 2002. Mathematical models for local nontexture inpaintings. SIAM Journal on Applied Mathematics 62(3), pp. 1019--1043

    2002

  39. [47]

    C., Sousa, D., Bispo, A

    Siravenha, A. C., Sousa, D., Bispo, A. and Pelaes, E., 2011. Evaluating inpainting methods to the satellite images clouds and shadows removing. In: Signal Processing, Image Processing and Pattern Recognition: International Conference, SIP 2011, Held as Part of the Future Gener...

    2011

  40. [48]

    and Li, Z., 2023

    Wang, Y., Zhang, B., Zhang, W., Hong, D., Zhao, B. and Li, Z., 2023. Cloud removal with sar-optical data fusion using a unified spatial-spectral residual network. IEEE Transactions on Geoscience and Remote Sensing

    2023

  41. [49]

    and Yu, W.-x., 2008

    Wu, Y.-h., Ji, K.-f. and Yu, W.-x., 2008. Comparison of classification performance of full-, dual-and single-polarization sar images using svm. Journal of Remote Sensing 12(1), pp. 53

    2008

  42. [50]

    and Zhu, X

    Xu, F., Shi, Y., Ebel, P., Yu, L., Xia, G.-S., Yang, W. and Zhu, X. X., 2022. Glf-cr: Sar-enhanced cloud removal with global--local fusion. ISPRS Journal of Photogrammetry and Remote Sensing 192, pp. 268--278

    2022

  43. [51]

    and Yamada, H., 2005

    Yamaguchi, Y., Moriyama, T., Ishido, M. and Yamada, H., 2005. Four-component scattering model for polarimetric sar image decomposition. IEEE Transactions on geoscience and remote sensing 43(8), pp. 1699--1706

    2005

  44. [52]

    and Yamada, H., 2011

    Yamaguchi, Y., Sato, A., Boerner, W.-M., Sato, R. and Yamada, H., 2011. Four-component scattering power decomposition with rotation of coherency matrix. IEEE Transactions on Geoscience and Remote Sensing 49(6), pp. 2251--2258

    2011

  45. [53]

    and Travis, D

    Zhang, C., Li, W. and Travis, D. J., 2009. Restoration of clouded pixels in multispectral remotely sensed imagery with cokriging. International Journal of Remote Sensing 30(9), pp. 2173--2195

    2009

  46. [54]

    and Deng, L., 2020

    Zhang, C., Yang, Z., He, X. and Deng, L., 2020. Multimodal intelligence: Representation learning, information fusion, and applications. IEEE Journal of Selected Topics in Signal Processing 14(3), pp. 478--493

    2020

  47. [55]

    and Jin, Y.-Q., 2017

    Zhang, Z., Wang, H., Xu, F. and Jin, Y.-Q., 2017. Complex-valued convolutional neural network and its application in polarimetric sar image classification. IEEE Transactions on Geoscience and Remote Sensing 55(12), pp. 7177--7188

    2017

  48. [56]

    and Yang, M.-H., 2021

    Zhou, J., Jampani, V., Pi, Z., Liu, Q. and Yang, M.-H., 2021. Decoupled dynamic filter networks. In: Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, pp. 6647--6656

    2021

  49. [57]

    and Ren, J., 2019

    Zhou, S., Zhang, J., Pan, J., Xie, H., Zuo, W. and Ren, J., 2019. Spatio-temporal filter adaptive network for video deblurring. In: Proceedings of the IEEE/CVF international conference on computer vision, pp. 2482--2491

    2019

  50. [58]

    and Jin, Y.-Q., 2016

    Zhou, Y., Wang, H., Xu, F. and Jin, Y.-Q., 2016. Polarimetric sar image classification using deep convolutional neural networks. IEEE Geoscience and Remote Sensing Letters 13(12), pp. 1935--1939

    2016

Pith tools

Reviewed August 10, 2026 · model on record in the stance chip above.