Pith. sign in

REVIEW 3 major objections 5 minor 41 references

A Comparative Study of U-Net Architectures for Change Detection in Satellite Images

T0 review · 3 major / 5 minor · reviewed 2026-08-07 · deepseek-v4-flash

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

Pith's one-line read This paper compiles reported results for 18 U-Net variants and argues that design choices such as Siamese branches, transformers, and attention gates address specific U-Net limitations in satellite change detection.

desk verdict A survey that promises more than it delivers: the categorization is useful, but the quantitative comparison mixes incomparable numbers and the abstract's 18-variant claim is never backed up. read the letter →

arxiv 2506.07925 v1 pith:V7LJAVED submitted 2025-06-09 cs.CV cs.LGeess.IV

Yaxita Amin , Naimisha S Trivedi , Rashmi Bhattad This is my paper
classification cs.CVcs.LGeess.IV
keywords remotesensingchangedetectionU-NetsatelliteimageryimagesegmentationattentionmechanismSwintransformerdeeplearning
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 surveys the growing family of U-Net-based architectures for detecting landscape changes in satellite imagery. It catalogs 18 variants, groups their modifications into attention mechanisms, transformer blocks, transfer learning, and skip-connection refinements, and assembles reported performance scores from 34 papers. The aim is to tell practitioners which design choices matter for remote sensing change detection, where images from different times and sensors must be compared. The headline result is a ranking in which NDR-U-Net, Siamese Swin-U-Net, and Optimised U-Net lead on different metrics. A sympathetic reader would take the paper as a structured map of the design space, not as a controlled benchmark.

What carries the argument

The central object is the U-Net variant taxonomy, organized around five modification strategies: attention gates, transformer blocks, transfer-learning encoders, encoder or skip-connection redesigns, and hierarchical nested blocks. The mechanism that carries the argument is the mapping from U-Net components to known failure modes: the encoder to receptive field, the decoder to global context, and skip connections to feature fusion. Each variant is read as an intervention on one component, and reported metrics (F1, Mean IoU, Overall Accuracy) are used as evidence for which interventions help.

What would settle it

Run NDR-U-Net, Siamese Swin-U-Net, and Optimised U-Net on one shared benchmark, such as LEVIR-CD, with identical preprocessing, training protocol, and metric computation; the paper's ranking stands only if their relative order under these controlled scores matches Table 4.

Watch

Extended reading notes

Core claim

This paper claims that the standard U-Net, inherited from medical segmentation, stumbles on satellite change detection for three reasons: limited receptive field, weak global context, and skip connections that merge semantically distant features. It then shows how 18 variants target these weaknesses: Swin transformers and attention modules capture long-range dependencies, pre-trained encoders like EfficientNetV2 bring transfer learning, bottleneck and bilateral attention refine skip connections, and nested dense residual blocks add multi-scale features. Based on assembled scores, the paper identifies NDR-U-Net (Mean IoU 0.92) as the leading segmentation result and Siamese Swin-U-Net (F1 94.67) as the leading change detection classifier. The intended contribution is a practical map: choose a variant by matching its modification to the dominant failure mode of the task.

Load-bearing premise

The paper's ranking assumes that F1, Mean IoU, and Overall Accuracy scores taken from different publications, measured on different datasets and sensors, can be compared directly as if from a single benchmark.

Editorial extensions

If this is right

  • If the assembled scores are taken at face value, NDR-U-Net is the variant to try first for segmentation-heavy change detection, since it reports the highest Mean IoU (0.92).
  • Siamese Swin-U-Net's F1 of 94.67 makes it the leading candidate when the task is binary land-cover change classification.
  • The paper's mapping from U-Net components to failure modes implies that skip-connection fixes (bottlenecks, bilateral attention) and long-range modules (transformers) address different problems, so neither alone covers all cases.
  • Practitioners with small datasets should favor transfer-learning variants such as STCD-EffV2T U-Net, since pre-trained encoders are the paper's stated remedy for limited training data.

Reading between the lines

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

  • Beyond the paper: a controlled re-run of the top variants on one benchmark would be needed before treating the Table 4 ranking as a leaderboard, since the paper compiles scores from different studies rather than running a single experiment.
  • Beyond the paper: the taxonomy suggests a natural next test, combining a Swin-transformer encoder with an attention-gated skip connection, because those two modifications target the receptive-field and skip-connection limitations separately.
  • Beyond the paper: because the datasets span optical, LiDAR, and different resolutions, the paper's data hint that sensor-specific adapter modules may matter more than any single architecture choice, though the paper does not test this.
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 / 5 minor

Summary. This manuscript presents a narrative review of U-Net architecture variants applied to remote sensing change detection. It surveys 34 papers, categorizes eight U-Net variants (Siamese Swin-U-Net, STCD-EffV2T U-Net, T-U-Net, Ensemble U-Net-ResNet, Optimised U-Net, Bilateral Attention U-Net, NDR-U-Net, HARNU-Net), and tabulates their reported performance scores. The stated goal is to help practitioners choose among U-Net variants for change detection, with conclusions such as 'NDR-U-Net achieves the highest Mean IoU (0.92)' and that Optimised U-Net and NDR-U-Net are 'top contenders.' The paper also discusses limitations of standard U-Net and qualitatively maps each variant to mitigation strategies for remote sensing change detection challenges.

Significance. If the comparative assessment were valid, it would serve as a useful entry point for practitioners selecting U-Net backbones for change detection. The paper compiles a relevant set of recent architectures and identifies plausible design themes (attention mechanisms, transfer learning, skip-connection modifications, transformer blocks). It also makes the worthwhile observation that standard U-Net's limited receptive field and coarse skip connections are bottlenecks for remote sensing change detection. However, the paper's central contribution is a ranking of variants based on incomparable published scores, and the lack of any controlled evaluation or explicit inclusion criteria undermines this ranking. As a review, the paper would require a systematic methodology and a transparent statement of what can and cannot be concluded from heterogeneous literature values, neither of which is present.

major comments (3)
  1. [Section 4.1, Table 4] Table 4 is the sole empirical basis for the comparative claim, but it juxtaposes performance scores measured on different datasets, sensors, resolutions, and metrics. For example, Siamese Swin-U-Net reports F1=94.67 on CDD, STCD-EffV2T U-Net reports Mean IoU=0.87 on OSCD/Sentinel-2, and NDR-U-Net reports Mean IoU=0.92 on GF-2. These numbers do not control for dataset difficulty, class balance, evaluation protocol, or even the same metric. Without a common benchmark or error bars, the statement that 'NDR-U-Net achieves the highest Mean IoU (0.92)' and the designation of 'top contenders' are not justified. This incomparability is load-bearing: if it is not addressed, the paper's central conclusion collapses into a list of reported values.
  2. [Abstract and Section 5] The abstract claims 'a comparison and analysis of 18 different U-Net variations,' but Section 5 states 'we study 8 variants,' and Tables 2, 3, and 4 list only 8 architectures. The discrepancy is never explained, and no inclusion or exclusion criteria are given for the 34 papers surveyed. Consequently, the reader cannot assess whether the selected variants and papers are representative, which is a second load-bearing weakness for a review whose stated purpose is to provide a comparative assessment.
  3. [Section 3.2, Table 2] Several architecture descriptions are inconsistent with the cited sources or with the paper's own table. For instance, Table 2 lists T-U-Net as 'Triple branch architecture,' but the text describes it as incorporating a transformer module in the decoder; both T-U-Net and Optimised U-Net are cited to the same reference [29], which appears to be the Optimised U-Net paper only. Additionally, the description of NDR-U-Net is hedged as 'it likely introduces modifications in the encoder path,' which is speculative rather than a verified account of the cited method. These inconsistencies impair the accuracy of the qualitative comparison that the paper does provide.
minor comments (5)
  1. [Abstract] The abstract contains grammatical errors and typos, such as 'this paper fill the gap' and 'ever changing' (missing hyphen), which should be corrected.
  2. [Figure 1] Figure 1 is referenced as showing a 'substantial increase' in U-Net variant publications, but the figure appears to be a line chart without axes labels or a data source; the claim is not substantiated by the displayed content.
  3. [Section 3.1] The description of Stack U-Net (reference [13]) is vague ('Stack U-Net incorporates multi-resolution feature extraction'), and the reference list entry lacks full publication details, making it difficult to locate the original work.
  4. [Section 4.1] The sentence 'Optimized U-Net shows promising overall accuracy (92.10%) for land cover classification' is presented without context on the dataset or the number of classes, which would be needed for any meaningful interpretation.
  5. [Section 4.2] The discussion of techniques such as min-max normalization and data augmentation is not tied to any specific variant or experimental result, making these paragraphs loosely connected to the paper's comparative theme.

Circularity Check

0 steps flagged · score 0.0 of 10

No circularity; the paper is a review that reports external results rather than deriving predictions from its own fitted inputs.

full rationale

The paper is a comparative literature review of U-Net variants for remote sensing change detection. It does not introduce a method, fit parameters, or derive a result from an ansatz. The central claim that NDR-U-Net achieves the highest Mean IoU (0.92) is a direct transcription of a performance number reported in an external cited paper, and the comparison table is assembled from published scores. There are no self-citations among the authors, no uniqueness theorem invoked from prior work by the same authors, and no fitted parameter renamed as a prediction. The internal inconsistency between the abstract's '18 different U-Net variations'/'34 papers' and Section 5's 'we study 8 variants' is a reporting or scope error, not circular reasoning. The incomparability of metrics across different datasets and protocols is a validity threat to the comparative conclusion, but it is not a circularity: the paper does not define its conclusion into existence; it simply reads the largest number from an unnormalized table. Therefore, no load-bearing step reduces to its own inputs.

Assumptions & free parameters 0 free parameters · 2 assumptions · 0 invented entities

The paper introduces no free parameters or invented entities. It rests on the assumptions that the surveyed papers are representative and that their differing evaluation setups allow cross-model comparison. Both assumptions are stated nowhere and are questionable.

assumptions (2)
  • domain assumption Reported scores across different datasets, sensors, resolutions, and metrics can be compared to rank U-Net variants.
    Invoked in Table 4 and Section 4.1, where F1, Mean IoU, and Overall Accuracy from different papers are listed together and NDR-U-Net is declared best.
  • ad hoc to paper The set of 34 papers and 18 variants is representative of the field, despite no systematic search or inclusion criteria being described.
    Sections 3 and 4 introduce the variant list without stating how it was compiled; the abstract's claim of 18 variants conflicts with the 8 detailed in the body.

how reviews work

0 comments
Cite this review

Pith. "Pith review of A Comparative Study of U-Net Architectures for Change Detection in Satellite Images." pith.science (2026). https://pith.science/paper/V7LJAVED

@misc{pith2026250607925,
  author       = {Pith},
  title        = {Pith review of: A Comparative Study of U-Net Architectures for Change Detection in Satellite Images},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/V7LJAVED}},
  note         = {Machine review of arXiv:2506.07925}
}
read the original abstract

Remote sensing change detection is essential for monitoring the everchanging landscapes of the Earth. The U-Net architecture has gained popularity for its capability to capture spatial information and perform pixel-wise classification. However, their application in the Remote sensing field remains largely unexplored. Therefore, this paper fill the gap by conducting a comprehensive analysis of 34 papers. This study conducts a comparison and analysis of 18 different U-Net variations, assessing their potential for detecting changes in remote sensing. We evaluate both benefits along with drawbacks of each variation within the framework of this particular application. We emphasize variations that are explicitly built for change detection, such as Siamese Swin-U-Net, which utilizes a Siamese architecture. The analysis highlights the significance of aspects such as managing data from different time periods and collecting relationships over a long distance to enhance the precision of change detection. This study provides valuable insights for researchers and practitioners that choose U-Net versions for remote sensing change detection tasks.

Discussion (0). Sign in to comment.

Reference graph

Works this paper leans on

41 extracted references · 33 canonical work pages

  1. [29]

    Concurrent Spatial and Channel Squeeze & Excitation in Fully Convolutional Networks,

    A. G. Roy, N. Navab, and C. Wachinger, “Concurrent Spatial and Channel Squeeze & Excitation in Fully Convolutional Networks,” 2018

    work page 2018

  2. [1]

    A review on image segmentation techniques with remote sensing perspective,

    V. Dey, Y. Zhang, and M. Zhong, “A review on image segmentation techniques with remote sensing perspective,” ISPRS TC VII Symposium - 100 Years ISPRS, vol. 38, 2010

    work page 2010

  3. [2]

    fully convolutional network

    Background Remote sensing change detection provides a rich set of functionalities as a development that is constantly evolving. A wide spectrum of techniques is available, however, deep learning methods are still better in many ways due to their effective learning capability and adapt to diverse types of data [9].Following the development of U-Net: Convol...

    work page

  4. [3]

    By utilizing knowledge from large dataset, U-Net variants can achieve high performance with less training data

    Transfer Learning U-Net Variants: Transfer learning is powerful technique for leveraging pre -trained models in remote sensing tasks. By utilizing knowledge from large dataset, U-Net variants can achieve high performance with less training data. In transfer learning, crucial step is to choose right pre -trained model suitable to remote sensing task. The p...

    work page

  5. [4]

    These variants introduce modifications within the encoder path to enhance feature extraction for change detection tasks

    U-Nets with Specific Encoder Modifications: In encoder path, baseline U-Net struggles to capture complex feature for task like VHR/HR images in remote sens ing. These variants introduce modifications within the encoder path to enhance feature extraction for change detection tasks. T-U-Net [29] This variant integrates a transformer module in the decoder pa...

    work page

  6. [5]

    For example, Optimised U-Net [29] ,introduces bottleneck structures within skip connections

    U-Nets with Specific Skip Connection Modifications: it focus on refining the information flow within skip connections to improve change detection accuracy. For example, Optimised U-Net [29] ,introduces bottleneck structures within skip connections. Skip connections can struggle with dimensionality issues when combining features from different resolutions....

    work page

  7. [6]

    With 3 main components of U- Net, they are trying to modify and manipulate among those parts

    Other modifications: For boosting accuracy of baseline U-Net, many researchers are modifying on and upon baseline U-Net architecture. With 3 main components of U- Net, they are trying to modify and manipulate among those parts. NDR-U-Net [32]: it likely introduces modifications in the encoder path. It might utilize multi -dilated convolutional layers to c...

    work page

  8. [7]

    A Unified Approach to Change Detection Using an Adaptive Ensemble of Extreme 8 Learning Machines,

    M. Khurana and V. Saxena, “A Unified Approach to Change Detection Using an Adaptive Ensemble of Extreme 8 Learning Machines,” IEEE Geoscience and Remote Sensing Letters, vol. PP, pp. 1–5, 2019, doi: 10.1109/LGRS.2019.2933906

    work page arXiv 2019

Show all 41 references
  1. [8]

    U-Net: Convolutional Networks for Biomedical Image Segmentation,

    O. Ronneberger, P. Fischer, and T. Brox, “U-Net: Convolutional Networks for Biomedical Image Segmentation,” LNCS, vol. 9351, pp. 234–241, 2015, doi: 10.1007/978-3-319-24574-4_28

    2015 doi

  2. [9]

    Review Article Digital Change Detection Techniques Using Remotely-Sensed Data,

    A. Singh, “Review Article Digital Change Detection Techniques Using Remotely-Sensed Data,” International Journal of Remote Sensing, vol. 10, no. 6, pp. 989–1003, 1989, doi: 10.1080/01431168908903939

    1989 doi

  3. [10]

    Change Detection Based on Deep Siamese Convolutional Network for Optical Aerial Images,

    Y. Zhan, K. Fu, M. Yan, X. Sun, H. Wang, and X. Qiu, “Change Detection Based on Deep Siamese Convolutional Network for Optical Aerial Images,” IEEE Geoscience and Remote Sensing Letters, vol. PP, pp. 1–5, 2017, doi: 10.1109/LGRS.2017.2738149

    2017

  4. [11]

    Densely Connected Convolutional Networks,

    G. Huang, Z. Liu, and K. Q. Weinberger, “Densely Connected Convolutional Networks,” 2017 IEEE Conference on Computer Vision and Pattern Recognition (CVPR), pp. 2261–2269, 2016

    2017

  5. [12]

    A Deep Convolutional Coupling Network for Change Detection Based on Heterogeneous Optical and Radar Images,

    J. Liu, M. Gong, K. Qin, and P. Zhang, “A Deep Convolutional Coupling Network for Change Detection Based on Heterogeneous Optical and Radar Images,” IEEE Transactions on Neural Networks and Learning Systems, vol. 29, no. 3, pp. 545–559, 2018, doi: 10.1109/TNNLS.2016.2636227

    2018

  6. [13]

    Toward Generalized Change Detection on Planetary Surfaces With Convolutional Autoencoders and Transfer Learning,

    H. R. Kerner, K. L. Wagstaff, B. D. Bue, P. C. Gray, J. F. Bell III, and H. B. Amor, “Toward Generalized Change Detection on Planetary Surfaces With Convolutional Autoencoders and Transfer Learning,” IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sens...

    2019

  7. [14]

    U- Net Transformer: Self and Cross Attention for Medical Image Segmentation,

    O. Petit, N. Thome, C. Rambour, and L. Soler, “U- Net Transformer: Self and Cross Attention for Medical Image Segmentation,” ArXiv, vol. abs/2103.06104, 2021

    2021 arXiv

  8. [15]

    Going Deeper with Convolutions,

    C. Szegedy et al., “Going Deeper with Convolutions,” 2015 IEEE Conference on Computer Vision and Pattern Recognition (CVPR), pp. 1–9, 2014

    2015

  9. [16]

    Deep Learning-Based Change Detection in Remote Sensing Images: A Review,

    A. Shafique, G. Cao, Z. Khan, M. Asad, and M. Aslam, “Deep Learning-Based Change Detection in Remote Sensing Images: A Review,” Remote Sensing, vol. 14, no. 4, p. 871, 2022, doi: 10.3390/rs14040871

    2022 doi

  10. [17]

    HARNU-Net: Hierarchical Attention Residual Nested U-Net for Change Detection in Remote Sensing Images,

    H. Li, L. Wang, and S. Cheng, “HARNU-Net: Hierarchical Attention Residual Nested U-Net for Change Detection in Remote Sensing Images,” Sensors, vol. 22, no. 12, p. 4626, 2022, doi: 10.3390/s22124626

    2022 doi

  11. [18]

    SegNet: A Deep Convolutional Encoder-Decoder Architecture for Image Segmentation,

    V. Badrinarayanan, A. Kendall, and R. Cipolla, “SegNet: A Deep Convolutional Encoder-Decoder Architecture for Image Segmentation,” IEEE Transactions on Pattern Analysis and Machine Intelligence, vol. 39, no. 12, pp. 2481–2495, 2017, doi: 10.1109/TPAMI.2016.2644615

    2017

  12. [19]

    Dense-UNet: A novel multiphoton in vivo cellular image segmentation model based on a convolutional neural network,

    S. Cai, Y. Tian, H. Lui, H. Zeng, Y. Wu, and G. Chen, “Dense-UNet: A novel multiphoton in vivo cellular image segmentation model based on a convolutional neural network,” Quantitative Imaging in Medicine and Surgery, vol. 10, no. 6, pp. 1275–1285, 2020, doi: 10.21037/qims-19- 1090

    2020 doi

  13. [20]

    Stacked U-Nets: A No-Frills Approach to Natural Image Segmentation,

    D. Shah, P. Ghosh, L. Davis, and T. Goldstein, “Stacked U-Nets: A No-Frills Approach to Natural Image Segmentation,” 2018

    2018

  14. [21]

    Recalibrating Fully Convolutional Networks with Spatial and Channel ‘Squeeze & Excitation’ Blocks,

    A. G. Roy, N. Navab, and C. Wachinger, “Recalibrating Fully Convolutional Networks with Spatial and Channel ‘Squeeze & Excitation’ Blocks,” 2018

    2018

  15. [22]

    Road Extraction by Deep Residual U-Net,

    Z. Zhang, Q. Liu, and Y. Wang, “Road Extraction by Deep Residual U-Net,” IEEE Geoscience and Remote Sensing Letters, vol. 15, no. 5, pp. 749–753, 2018, doi: 10.1109/LGRS.2018.2802944

    2018

  16. [23]

    Recurrent residual U-Net for medical image segmentation,

    M. Z. Alom, C. Yakopcic, M. Hasan, T. M. Taha, and V. K. Asari, “Recurrent residual U-Net for medical image segmentation,” Journal of Medical Imaging (Bellingham), vol. 6, no. 1, p. 014006, 2019, doi: 10.1117/1.JMI.6.1.014006

    2019 doi

  17. [24]

    Attention U-Net: Learning Where to Look for the Pancreas,

    O. Oktay et al., “Attention U-Net: Learning Where to Look for the Pancreas,” 2018

    2018

  18. [25]

    Based on the calculated similarity score, the attention module generates weight map

    it focuses on the relationship between pixels, captures local dependencies and channel attention, it gives importance to channels within feature maps. Based on the calculated similarity score, the attention module generates weight map. Siamese Swin [26]. It replaces standard c...

  19. [26]

    UNet++: A Nested U-Net Architecture for Medical Image Segmentation,

    Z. Zhou, M. M. Rahman Siddiquee, N. Tajbakhsh, and J. Liang, “UNet++: A Nested U-Net Architecture for Medical Image Segmentation,” 4th International Workshop, DLMIA 2018, and 8th International Workshop, ML-CDS 2018, Held in Conjunction with MICCAI 2018, vol. 11045, pp. 3–11, 2...

    2018 doi

  20. [27]

    It allows the network to focus on informative features

    integrates bilateral attention mechanisms within skip connections. It allows the network to focus on informative features. Where HARNU-Net [10] addressed limitations like limited receptive field and skip connection issues. Architecture Name Evaluation Metric Performance Score ...

  21. [28]

    USE-Net: Incorporating Squeeze- and-Excitation Blocks into U-Net for Prostate Zonal Segmentation of Multi-Institutional MRI Datasets,

    L. Rundo et al., “USE-Net: Incorporating Squeeze- and-Excitation Blocks into U-Net for Prostate Zonal Segmentation of Multi-Institutional MRI Datasets,” Neurocomputing, vol. 365, pp. 31–43, 2019

    2019

  22. [30]

    Deep Residual Learning for Image Recognition,

    K. He, X. Zhang, S. Ren, and J. Sun, “Deep Residual Learning for Image Recognition,” 2016 IEEE Conference on Computer Vision and Pattern Recognition (CVPR), pp. 770–778, 2016, doi: 10.1109/CVPR.2016.90

    2016 doi

  23. [31]

    Multi-Class U- Net for Segmentation of Non-Biometric Identifiers,

    I. Hrkać, H. Brkić, and I. Kalafatić, “Multi-Class U- Net for Segmentation of Non-Biometric Identifiers,” The 12th International Conference on Computer Vision and Image Processing (IMVIP 2017), pp. 7–12, 2017

    2017

  24. [32]

    SA-UNet: Spatial Attention U-Net for Retinal Vessel Segmentation,

    C. Guo, M. Szemenyei, Y. Yi, W. Wang, B. Chen, and C. Fan, “SA-UNet: Spatial Attention U-Net for Retinal Vessel Segmentation,” 2020

    2020

  25. [33]

    A Siamese Swin-Unet for Image Change Detection,

    Y. Tang, Z. Cao, N. Guo, and M. Jiang, “A Siamese Swin-Unet for Image Change Detection,” Scientific Reports, vol. 14, no. 1, p. 4577, 2024, doi: 10.1038/s41598-024- 54096-8

    2024 doi

  26. [34]

    Bilateral Attention U-Net with Dissimilarity Attention Gate for Change Detection on Remote Sensing Imageries,

    J. Lee, W. Wiratama, W. Lee, I. Marzuki, and D. Sim, “Bilateral Attention U-Net with Dissimilarity Attention Gate for Change Detection on Remote Sensing Imageries,” Applied Sciences, vol. 13, no. 4, p. 2485, 2023, doi: 10.3390/app13042485

    2023 doi

  27. [35]

    STCD- EffV2T U-Net: Semi Transfer Learning EfficientNetV2 T-U- Net Network for Urban/Land Cover Change Detection Using Sentinel-2 Satellite Images,

    M. Gomroki, M. Hasanlou, and P. Reinartz, “STCD- EffV2T U-Net: Semi Transfer Learning EfficientNetV2 T-U- Net Network for Urban/Land Cover Change Detection Using Sentinel-2 Satellite Images,” Remote Sensing, vol. 15, no. 5, p. 1232, 2023, doi: 10.3390/rs15051232. 9

    2023 doi

  28. [36]

    Optimised U-Net for Land Use–Land Cover Classification Using Aerial Photography,

    A. Clark, S. Phinn, and P. Scarth, “Optimised U-Net for Land Use–Land Cover Classification Using Aerial Photography,” PFG – Journal of Photogrammetry Remote Sensing and Geoinformation Science, vol. 91, pp. 1–23, 2023, doi: 10.1007/s41064-023-00233-3

    2023 doi

  29. [37]

    Batch Normalization: Accelerating Deep Network Training by Reducing Internal Covariate Shift,

    S. Ioffe and C. Szegedy, “Batch Normalization: Accelerating Deep Network Training by Reducing Internal Covariate Shift,” 2015

    2015

  30. [38]

    Adam: A Method for Stochastic Optimization,

    D. Kingma and J. Ba, “Adam: A Method for Stochastic Optimization,” International Conference on Learning Representations, 2014

    2014

  31. [39]

    NDR-UNet: Segmentation of Water Bodies in Remote Sensing using Nested Dense Residual U-Net,

    J. Surekha and N. Gupta, “NDR-UNet: Segmentation of Water Bodies in Remote Sensing using Nested Dense Residual U-Net,” 2024, doi: 10.21203/rs.3.rs- 3133244/v1

    2024 doi

  32. [40]

    Change Detection Methods for Remote Sensing in the Last Decade: A Comprehensive Review,

    G. Cheng et al., “Change Detection Methods for Remote Sensing in the Last Decade: A Comprehensive Review,” 2023

    2023

  33. [41]

    Advances and Challenges in Deep Learning-Based Change Detection for Remote Sensing Images: A Review Through Various Learning Paradigms,

    L. Wang, M. Zhang, X. Gao, and W. Shi, “Advances and Challenges in Deep Learning-Based Change Detection for Remote Sensing Images: A Review Through Various Learning Paradigms,” Remote Sensing, vol. 16, no. 5, p. 804, 2024, doi: 10.3390/rs16050804

    2024 doi

Pith tools

Reviewed August 7, 2026 · model on record in the stance chip above.