ISPRS Journal of Photogrammetry and Remote Sensing最新文献_第9页

Optimizing hybrid models for canopy nitrogen mapping from Sentinel-2 in Google Earth Engine 在谷歌地球引擎中优化利用哨兵 2 号卫星绘制冠层氮含量图的混合模型

IF 10.6 1区地球科学 Q1 GEOGRAPHY, PHYSICAL

ISPRS Journal of Photogrammetry and Remote Sensing

Pub Date : 2024-11-22 DOI: 10.1016/j.isprsjprs.2024.11.005

Emma De Clerck , Dávid D.Kovács , Katja Berger , Martin Schlerf , Jochem Verrelst

Canopy nitrogen content (CNC) is a crucial variable for plant health, influencing photosynthesis and growth. An optimized, scalable approach for spatially explicit CNC quantification using Sentinel-2 (S2) data is presented, integrating PROSAIL-PRO simulations with Gaussian Process Regression (GPR) and an Active Learning technique, specifically the Euclidean distance-based diversity (EBD) approach for selective sampling. This hybrid method enhances training dataset efficiency and optimizes CNC models for practical applications. Two GPR models based on PROSAIL-PRO variables were evaluated: a protein-based model (C

_{prot}

-LAI) and a chlorophyll-based model (C

_{ab}

-LAI). Both models, implemented in Google Earth Engine (GEE), demonstrated strong performance and outperformed other machine learning methods, including kernel ridge regression, principal component regression, neural network, weighted k-nearest neighbors regression, partial least squares regression and least squares linear regression. Validation results showed moderate to good accuracies: NRMSE

_{C_{prot - LAI}}

= 16.76%, R

_{C_{prot - LAI}}^{2}

= 0.47; NRMSE

_{C_{ab - LAI}}

= 18.74%, R

_{C_{ab - LAI}}^{2}

= 0.51. The models revealed high consistency for an independent validation dataset of the Munich-North-Isar (Germany) test site, with R

^{2}

values of 0.58 and 0.71 and NRMSEs of 21.47% and 20.17% for the C

_{prot}

-LAI model and C

_{ab}

-LAI model, respectively. The models also demonstrated high consistency across growing seasons, indicating their potential for time series analysis of CNC dynamics. Application of the S2-based mapping workflow across the Iberian Peninsula, with estimates showing relative uncertainty below 30%, highlights the model’s broad applicability and portability. The optimized EBD-GPR-CNC approach within GEE supports scalable CNC estimation and offers a robust tool for monitoring nitrogen dynamics.

冠层氮含量（CNC）是植物健康的关键变量，影响着光合作用和生长。本文介绍了一种利用哨兵-2（Sentinel-2，S2）数据进行空间显式氮含量量化的优化、可扩展方法，该方法将 PROSAIL-PRO 模拟与高斯过程回归（GPR）和主动学习技术（特别是用于选择性采样的基于欧氏距离的多样性（EBD）方法）相结合。这种混合方法提高了训练数据集的效率，优化了 CNC 模型的实际应用。评估了两个基于 PROSAIL-PRO 变量的 GPR 模型：一个基于蛋白质的模型（Cprot-LAI）和一个基于叶绿素的模型（Cab-LAI）。这两个模型都是在谷歌地球引擎（GEE）中实现的，表现出色，优于其他机器学习方法，包括核岭回归、主成分回归、神经网络、加权 k 近邻回归、偏最小二乘回归和最小二乘线性回归。验证结果显示了中等到良好的精确度：NRMSECprot-LAI = 16.76%，RCprot-LAI2 = 0.47；NRMSECab-LAI = 18.74%，RCab-LAI2 = 0.51。这些模型在慕尼黑-北伊萨尔（德国）试验场的独立验证数据集上显示出高度一致性，Cprot-LAI 模型和 Cab-LAI 模型的 R2 值分别为 0.58 和 0.71，NRMSE 分别为 21.47% 和 20.17%。这些模型在不同的生长季节也表现出高度的一致性，表明它们具有对 CNC 动态进行时间序列分析的潜力。基于 S2 的绘图工作流程在伊比利亚半岛的应用，估算结果显示相对不确定性低于 30%，这突出表明了该模型的广泛适用性和可移植性。在 GEE 中优化的 EBD-GPR-CNC 方法支持可扩展的 CNC 估算，为监测氮动态提供了一个强大的工具。

{"title":"Optimizing hybrid models for canopy nitrogen mapping from Sentinel-2 in Google Earth Engine","authors":"Emma De Clerck , Dávid D.Kovács , Katja Berger , Martin Schlerf , Jochem Verrelst","doi":"10.1016/j.isprsjprs.2024.11.005","DOIUrl":"10.1016/j.isprsjprs.2024.11.005","url":null,"abstract":"<div><div>Canopy nitrogen content (CNC) is a crucial variable for plant health, influencing photosynthesis and growth. An optimized, scalable approach for spatially explicit CNC quantification using Sentinel-2 (S2) data is presented, integrating PROSAIL-PRO simulations with Gaussian Process Regression (GPR) and an Active Learning technique, specifically the Euclidean distance-based diversity (EBD) approach for selective sampling. This hybrid method enhances training dataset efficiency and optimizes CNC models for practical applications. Two GPR models based on PROSAIL-PRO variables were evaluated: a protein-based model (C<span><math><msub><mrow></mrow><mrow><mtext>prot</mtext></mrow></msub></math></span>-LAI) and a chlorophyll-based model (C<span><math><msub><mrow></mrow><mrow><mtext>ab</mtext></mrow></msub></math></span>-LAI). Both models, implemented in Google Earth Engine (GEE), demonstrated strong performance and outperformed other machine learning methods, including kernel ridge regression, principal component regression, neural network, weighted k-nearest neighbors regression, partial least squares regression and least squares linear regression. Validation results showed moderate to good accuracies: NRMSE<span><math><msub><mrow></mrow><mrow><msub><mrow><mi>C</mi></mrow><mrow><mtext>prot</mtext><mo>−</mo><mtext>LAI</mtext></mrow></msub></mrow></msub></math></span> = 16.76%, R<span><math><msubsup><mrow></mrow><mrow><msub><mrow><mi>C</mi></mrow><mrow><mtext>prot</mtext><mo>−</mo><mtext>LAI</mtext></mrow></msub></mrow><mrow><mn>2</mn></mrow></msubsup></math></span> = 0.47; NRMSE<span><math><msub><mrow></mrow><mrow><msub><mrow><mi>C</mi></mrow><mrow><mtext>ab</mtext><mo>−</mo><mtext>LAI</mtext></mrow></msub></mrow></msub></math></span> = 18.74%, R<span><math><msubsup><mrow></mrow><mrow><msub><mrow><mi>C</mi></mrow><mrow><mtext>ab</mtext><mo>−</mo><mtext>LAI</mtext></mrow></msub></mrow><mrow><mn>2</mn></mrow></msubsup></math></span> = 0.51. The models revealed high consistency for an independent validation dataset of the Munich-North-Isar (Germany) test site, with R<span><math><msup><mrow></mrow><mrow><mn>2</mn></mrow></msup></math></span> values of 0.58 and 0.71 and NRMSEs of 21.47% and 20.17% for the C<span><math><msub><mrow></mrow><mrow><mtext>prot</mtext></mrow></msub></math></span>-LAI model and C<span><math><msub><mrow></mrow><mrow><mtext>ab</mtext></mrow></msub></math></span>-LAI model, respectively. The models also demonstrated high consistency across growing seasons, indicating their potential for time series analysis of CNC dynamics. Application of the S2-based mapping workflow across the Iberian Peninsula, with estimates showing relative uncertainty below 30%, highlights the model’s broad applicability and portability. The optimized EBD-GPR-CNC approach within GEE supports scalable CNC estimation and offers a robust tool for monitoring nitrogen dynamics.</div></div>","PeriodicalId":50269,"journal":{"name":"ISPRS Journal of Photogrammetry and Remote Sensing","volume":"218 ","pages":"Pages 530-545"},"PeriodicalIF":10.6,"publicationDate":"2024-11-22","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142691119","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"地球科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

A unique dielectric constant estimation for lunar surface through PolSAR model-based decomposition 通过基于 PolSAR 模型的分解估算月球表面独特的介电常数

IF 10.6 1区地球科学 Q1 GEOGRAPHY, PHYSICAL

ISPRS Journal of Photogrammetry and Remote Sensing

Pub Date : 2024-11-22 DOI: 10.1016/j.isprsjprs.2024.10.022

Inderkumar Kochar , Anup Das , Rajib Kumar Panigrahi

Dielectric constant for the earth and planetary surfaces has been estimated using reflection coefficients in the past. A recent trend is to use model-based decomposition for dielectric constant retrieval from polarimetric synthetic aperture radar (polSAR) data. We examine the reported literature in this regard and propose a unique dielectric constant estimation (UDCE) algorithm using three-component decomposition technique. In UDCE, the dielectric constant is obtained directly from one of the elements of the measured coherency matrix in a single step. The dielectric constant estimate from the UDCE is independent of the volume scattering model when single-bounce or double-bounce scattering is dominant. This avoids error propagation from overestimation of volume scattering to the copolarization ratios, and in turn, to the dielectric constant, inherent in reported algorithms that use model-based decomposition. Consequently, a unique solution is obtained. We also demonstrate that the solution from the UDCE is unaffected by using a higher-order model-based decomposition. We evaluate the performance of the proposed UDCE algorithm over three Apollo 12, Apollo 15, and Apollo 17 landing sites on the lunar surface using Chandrayaan- 2 dual-frequency synthetic aperture radar (DFSAR) datasets. An excellent convergence rate for dielectric constant estimation is maintained over all three test sites. Using the proposed UDCE algorithm, the dielectric constant maps are produced for the lunar surface using full polSAR data for the first time. We observe that the generated dielectric constant maps capture all the ground truth features, previously unseen with such clarity.

地球和行星表面的介电常数过去一直使用反射系数来估算。最近的趋势是使用基于模型的分解方法从偏振合成孔径雷达（polSAR）数据中检索介电常数。我们研究了这方面的文献，提出了一种使用三分量分解技术的独特介电常数估计（UDCE）算法。在 UDCE 算法中，介电常数直接从测量到的相干矩阵的一个元素中一步获得。当单弹跳或双弹跳散射占主导地位时，UDCE 得出的介电常数估计值与体积散射模型无关。这就避免了高估体积散射对共极化比率，进而对介电常数造成的误差传播，而这种误差传播是已报道的使用基于模型分解的算法所固有的。因此，我们获得了唯一的解决方案。我们还证明，使用基于模型的高阶分解，UDCE 的解不会受到影响。我们使用 Chandrayaan- 2 双频合成孔径雷达 (DFSAR) 数据集对阿波罗 12 号、阿波罗 15 号和阿波罗 17 号在月球表面的三个着陆点评估了所提出的 UDCE 算法的性能。所有三个测试点的介电常数估计都保持了极佳的收敛率。利用所提出的 UDCE 算法，首次使用完整的 polSAR 数据生成了月球表面的介电常数图。我们观察到，生成的介电常数图捕捉到了所有地面实况特征，而这些特征以前从未如此清晰地呈现过。

{"title":"A unique dielectric constant estimation for lunar surface through PolSAR model-based decomposition","authors":"Inderkumar Kochar , Anup Das , Rajib Kumar Panigrahi","doi":"10.1016/j.isprsjprs.2024.10.022","DOIUrl":"10.1016/j.isprsjprs.2024.10.022","url":null,"abstract":"<div><div>Dielectric constant for the earth and planetary surfaces has been estimated using reflection coefficients in the past. A recent trend is to use model-based decomposition for dielectric constant retrieval from polarimetric synthetic aperture radar (polSAR) data. We examine the reported literature in this regard and propose a unique dielectric constant estimation (UDCE) algorithm using three-component decomposition technique. In UDCE, the dielectric constant is obtained directly from one of the elements of the measured coherency matrix in a single step. The dielectric constant estimate from the UDCE is independent of the volume scattering model when single-bounce or double-bounce scattering is dominant. This avoids error propagation from overestimation of volume scattering to the copolarization ratios, and in turn, to the dielectric constant, inherent in reported algorithms that use model-based decomposition. Consequently, a unique solution is obtained. We also demonstrate that the solution from the UDCE is unaffected by using a higher-order model-based decomposition. We evaluate the performance of the proposed UDCE algorithm over three Apollo 12, Apollo 15, and Apollo 17 landing sites on the lunar surface using Chandrayaan- 2 dual-frequency synthetic aperture radar (DFSAR) datasets. An excellent convergence rate for dielectric constant estimation is maintained over all three test sites. Using the proposed UDCE algorithm, the dielectric constant maps are produced for the lunar surface using full polSAR data for the first time. We observe that the generated dielectric constant maps capture all the ground truth features, previously unseen with such clarity.</div></div>","PeriodicalId":50269,"journal":{"name":"ISPRS Journal of Photogrammetry and Remote Sensing","volume":"218 ","pages":"Pages 546-554"},"PeriodicalIF":10.6,"publicationDate":"2024-11-22","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142691120","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"地球科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

Unwrap-Net: A deep neural network-based InSAR phase unwrapping method assisted by airborne LiDAR data Unwrap-Net：基于深度神经网络、由机载激光雷达数据辅助的 InSAR 相位解包方法

IF 10.6 1区地球科学 Q1 GEOGRAPHY, PHYSICAL

ISPRS Journal of Photogrammetry and Remote Sensing

Pub Date : 2024-11-21 DOI: 10.1016/j.isprsjprs.2024.11.009

Wang Yang , Yi He , Qing Zhu , Lifeng Zhang , Long Jin

In Interferometric Synthetic Aperture Radar (InSAR) data processing, accurately unwrapping the phase is crucial for measuring elevation or deformation. DCNN models such as PhaseNet and PGNet have improved the efficiency and accuracy of phase unwrapping, but they still face challenges such as incomplete multiscale feature learning, high feature redundancy, and reliance on unrealistic datasets. These limitations compromise their effectiveness in areas with low coherence and high gradient deformation. This study proposed Unwrap-Net, a novel network model featuring an encoder-decoder structure and enhanced multiscale feature learning via ASPP (Atrous Spatial Pyramid Pooling). Unwrap-Net minimizes feature redundancy and boosts learning efficiency using SERB (Residual Convolutional with SE-block). For dataset construction, airborne LiDAR terrain data combined with land cover data from optical images, using the SGS (Sequential Gaussian Simulation) method, are used to synthesize phase data and simulate decorrelation noise. This approach creates a dataset that closely approximates real-world conditions. Additionally, the introduction of a new high-fidelity optimization loss function significantly enhances the model’s resistance to noise. Experimental results show that compared to the SNAPHU and PhaseNet models, the SSIM of the Unwrap-Net model improves by over 13%, and the RMSE is reduced by more than 34% in simulated data experiments. In real data experiments, SSIM improves by over 6%, and RMSE is reduced by more than 49%. This indicates that the unwrapping results of the Unwrap-Net model are more reliable and have stronger generalization capabilities. The related experimental code and dataset will be made available at https://github.com/yangwangyangzi48/UNWRAPNETV1.git.

在干涉合成孔径雷达（InSAR）数据处理中，准确地解开相位对于测量高程或形变至关重要。PhaseNet 和 PGNet 等 DCNN 模型提高了相位解包的效率和准确性，但它们仍然面临着一些挑战，如多尺度特征学习不完整、特征冗余度高以及依赖不切实际的数据集。这些局限性影响了它们在低相干性和高梯度变形区域的有效性。本研究提出的 Unwrap-Net 是一种新型网络模型，具有编码器-解码器结构，并通过 ASPP（Atrous Spatial Pyramid Pooling）增强了多尺度特征学习。Unwrap-Net 利用 SERB（带 SE 块的残差卷积）将特征冗余最小化并提高学习效率。在数据集构建方面，采用 SGS（序列高斯模拟）方法将机载激光雷达地形数据与光学图像中的土地覆盖数据相结合，合成相位数据并模拟相关噪声。这种方法创建的数据集非常接近真实世界的条件。此外，新的高保真优化损失函数的引入大大增强了模型的抗噪能力。实验结果表明，在模拟数据实验中，与 SNAPHU 和 PhaseNet 模型相比，Unwrap-Net 模型的 SSIM 提高了 13% 以上，RMSE 降低了 34% 以上。在真实数据实验中，SSIM 提高了 6% 以上，RMSE 降低了 49% 以上。这表明，Unwrap-Net 模型的解包结果更加可靠，具有更强的泛化能力。相关实验代码和数据集将发布在 https://github.com/yangwangyangzi48/UNWRAPNETV1.git 网站上。

{"title":"Unwrap-Net: A deep neural network-based InSAR phase unwrapping method assisted by airborne LiDAR data","authors":"Wang Yang , Yi He , Qing Zhu , Lifeng Zhang , Long Jin","doi":"10.1016/j.isprsjprs.2024.11.009","DOIUrl":"10.1016/j.isprsjprs.2024.11.009","url":null,"abstract":"<div><div>In Interferometric Synthetic Aperture Radar (InSAR) data processing, accurately unwrapping the phase is crucial for measuring elevation or deformation. DCNN models such as PhaseNet and PGNet have improved the efficiency and accuracy of phase unwrapping, but they still face challenges such as incomplete multiscale feature learning, high feature redundancy, and reliance on unrealistic datasets. These limitations compromise their effectiveness in areas with low coherence and high gradient deformation. This study proposed Unwrap-Net, a novel network model featuring an encoder-decoder structure and enhanced multiscale feature learning via ASPP (Atrous Spatial Pyramid Pooling). Unwrap-Net minimizes feature redundancy and boosts learning efficiency using SERB (Residual Convolutional with SE-block). For dataset construction, airborne LiDAR terrain data combined with land cover data from optical images, using the SGS (Sequential Gaussian Simulation) method, are used to synthesize phase data and simulate decorrelation noise. This approach creates a dataset that closely approximates real-world conditions. Additionally, the introduction of a new high-fidelity optimization loss function significantly enhances the model’s resistance to noise. Experimental results show that compared to the SNAPHU and PhaseNet models, the SSIM of the Unwrap-Net model improves by over 13%, and the RMSE is reduced by more than 34% in simulated data experiments. In real data experiments, SSIM improves by over 6%, and RMSE is reduced by more than 49%. This indicates that the unwrapping results of the Unwrap-Net model are more reliable and have stronger generalization capabilities. The related experimental code and dataset will be made available at <span><span>https://github.com/yangwangyangzi48/UNWRAPNETV1.git</span><svg><path></path></svg></span>.</div></div>","PeriodicalId":50269,"journal":{"name":"ISPRS Journal of Photogrammetry and Remote Sensing","volume":"218 ","pages":"Pages 510-529"},"PeriodicalIF":10.6,"publicationDate":"2024-11-21","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142691121","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"地球科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

METNet: A mesh exploring approach for segmenting 3D textured urban scenes METNet：分割三维纹理城市场景的网格探索方法

IF 10.6 1区地球科学 Q1 GEOGRAPHY, PHYSICAL

ISPRS Journal of Photogrammetry and Remote Sensing

Pub Date : 2024-11-21 DOI: 10.1016/j.isprsjprs.2024.10.020

Qendrim Schreiber , Nicola Wolpert , Elmar Schömer

In this work, we present the neural network Mesh Exploring Tensor Net (METNet) for the segmentation of 3D urban scenes, that operates directly on textured meshes. Since triangular meshes have a very irregular structure, many existing approaches change the input by sampling evenly distributed point clouds on the meshes. The resulting simplified representation of the urban scenes has the advantage that state-of-the-art neural network architectures for point clouds can be used for the segmentation task. The disadvantages are that crucial geodesic information is lost and for a sufficiently good approximation of the scene many points are often necessary. Since memory on the GPU is limited, the consequence is that only small areas can be examined locally.

To overcome these limitations, METNet generates its input directly from the textured triangular mesh. It applies a new star-shaped exploration strategy, starting from a triangle on the mesh and expanding in various directions. This way, a vertex based regular local tensor from the unstructured mesh is generated, which we call a Mesh Exploring Tensor (MET). By expanding on the mesh, a MET maintains the information about the connectivity and distance of vertices along the surface of the mesh. It also effectively captures the characteristics of large regions. Our new architecture, METNet is optimized for processing METs as input. The regular structure of the input allows METNet the use of established convolutional neural networks.

Experimental results, conducted on two urban textured mesh benchmarks, demonstrate that METNet surpasses the performance of previous state-of-the-art techniques. METNet improves the previous state-of-the-art results by 8.6% in terms of mean IoU (intersection over union) on the SUM dataset compared to the second-best method PSSNet, and by 3.2% in terms of mean F1 score on the H3D dataset compared to the second-best method ifp-SCN. Our source code is available at https://github.com/QenSchr/METNet.

在这项工作中，我们介绍了用于三维城市场景分割的神经网络网格探索张量网（METNet），该网络可直接在纹理网格上运行。由于三角形网格具有非常不规则的结构，许多现有方法通过对网格上均匀分布的点云进行采样来改变输入。由此产生的城市场景简化表示法的优点是，最先进的点云神经网络架构可用于分割任务。缺点是会丢失关键的大地信息，而且为了对场景进行足够好的逼近，往往需要很多点。由于 GPU 的内存有限，因此只能对局部小区域进行检查。

{"title":"METNet: A mesh exploring approach for segmenting 3D textured urban scenes","authors":"Qendrim Schreiber , Nicola Wolpert , Elmar Schömer","doi":"10.1016/j.isprsjprs.2024.10.020","DOIUrl":"10.1016/j.isprsjprs.2024.10.020","url":null,"abstract":"<div><div>In this work, we present the neural network Mesh Exploring Tensor Net (METNet) for the segmentation of 3D urban scenes, that operates directly on textured meshes. Since triangular meshes have a very irregular structure, many existing approaches change the input by sampling evenly distributed point clouds on the meshes. The resulting simplified representation of the urban scenes has the advantage that state-of-the-art neural network architectures for point clouds can be used for the segmentation task. The disadvantages are that crucial geodesic information is lost and for a sufficiently good approximation of the scene many points are often necessary. Since memory on the GPU is limited, the consequence is that only small areas can be examined locally.</div><div>To overcome these limitations, METNet generates its input directly from the textured triangular mesh. It applies a new star-shaped exploration strategy, starting from a triangle on the mesh and expanding in various directions. This way, a vertex based regular local tensor from the unstructured mesh is generated, which we call a Mesh Exploring Tensor (MET). By expanding on the mesh, a MET maintains the information about the connectivity and distance of vertices along the surface of the mesh. It also effectively captures the characteristics of large regions. Our new architecture, METNet is optimized for processing METs as input. The regular structure of the input allows METNet the use of established convolutional neural networks.</div><div>Experimental results, conducted on two urban textured mesh benchmarks, demonstrate that METNet surpasses the performance of previous state-of-the-art techniques. METNet improves the previous state-of-the-art results by 8.6% in terms of mean IoU (intersection over union) on the SUM dataset compared to the second-best method PSSNet, and by 3.2% in terms of mean F1 score on the H3D dataset compared to the second-best method ifp-SCN. Our source code is available at <span><span>https://github.com/QenSchr/METNet</span><svg><path></path></svg></span>.</div></div>","PeriodicalId":50269,"journal":{"name":"ISPRS Journal of Photogrammetry and Remote Sensing","volume":"218 ","pages":"Pages 498-509"},"PeriodicalIF":10.6,"publicationDate":"2024-11-21","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142691123","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"地球科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

On-orbit geometric calibration of MERSI whiskbroom scanner MERSI 啸扫扫描仪的在轨几何校准

IF 10.6 1区地球科学 Q1 GEOGRAPHY, PHYSICAL

ISPRS Journal of Photogrammetry and Remote Sensing

Pub Date : 2024-11-18 DOI: 10.1016/j.isprsjprs.2024.11.007

Hongbo Pan, Xue Zhang, Zixuan Liu, Tao Huang

The whiskbroom scanner is a critical component in remote sensing payloads, such as the Visible Infrared Imaging Radiometer Suite (VIIRS) on the Joint Polar Satellite System (JPSS) and the Medium Resolution Spectral Imager (MERSI) on FengYun-3. However, panoramic distortion in whiskbroom scanner images increases overlap from the nadir to the edges between adjacent scans. These distortions present significant challenges for improving geolocation accuracy, particularly when errors occur in sensors and platforms. This manuscript derives analytic expressions for all potential error sources, including sensors, platforms, and elevation, using homogeneous coordinates in the focal plane. This derivation demonstrates that geolocation errors vary with view angles and detector positions. To further investigate these error properties, a gradient-aware least-squares matching method was developed to extract highly accurate and dense ground control points (GCPs) with approximately 100,000 points in a single scene. A three-step geometric calibration method was then introduced, which includes boresight misalignment correction, parametric geometric calibration, and non-uniform scanning compensation. Given the varying spatial resolution of the GCPs, the weight of the GCPs was dynamically updated for least-squares estimation. This method effectively demonstrated the complex geolocation errors in MERSI on FY-3D, a system that was not meticulously calibrated in the laboratory. The initial root mean square errors (RMSEs) were 3.354 and 12.441 instantaneous field of view (IFoV) for the designed parameters. The proposed geometric calibration method successfully corrected view-angle and detector position-related geolocation errors, reducing them to 0.211 and 0.225 IFoV in the scan and track directions, respectively. The geolocation validation software and experiment results were provided https://github.com/hongbop/whiskgeovalidation.git.

拂尘扫描仪是遥感有效载荷的一个重要组成部分，如联合极地卫星系统（JPSS）上的可见光红外成像辐射计套件（VIIRS）和风云三号上的中分辨率光谱成像仪（MERSI）。然而，拂尘扫描仪图像的全景失真会增加相邻扫描之间从天底到边缘的重叠。这些失真给提高地理定位精度带来了巨大挑战，特别是当传感器和平台出现误差时。本手稿利用焦平面上的均质坐标，推导出所有潜在误差源的分析表达式，包括传感器、平台和仰角。推导结果表明，地理定位误差随视角和探测器位置的变化而变化。为了进一步研究这些误差特性，我们开发了一种梯度感知最小二乘匹配方法，以提取高精度、高密度的地面控制点（GCP），单个场景中大约有 100,000 个点。然后引入了三步几何校准法，包括孔径偏差校正、参数几何校准和非均匀扫描补偿。鉴于 GCP 的空间分辨率不同，GCP 的权重被动态更新，以进行最小二乘估计。这种方法有效地展示了 FY-3D 上 MERSI 系统的复杂地理定位误差，而该系统并未在实验室中进行细致校准。设计参数的初始均方根误差（RMSE）分别为 3.354 和 12.441 瞬时视场（IFoV）。所提出的几何校准方法成功校正了与视角和探测器位置相关的地理定位误差，在扫描和跟踪方向将误差分别减小到 0.211 和 0.225 IFoV。地理定位验证软件和实验结果见 https://github.com/hongbop/whiskgeovalidation.git。

{"title":"On-orbit geometric calibration of MERSI whiskbroom scanner","authors":"Hongbo Pan, Xue Zhang, Zixuan Liu, Tao Huang","doi":"10.1016/j.isprsjprs.2024.11.007","DOIUrl":"10.1016/j.isprsjprs.2024.11.007","url":null,"abstract":"<div><div>The whiskbroom scanner is a critical component in remote sensing payloads, such as the Visible Infrared Imaging Radiometer Suite (VIIRS) on the Joint Polar Satellite System (JPSS) and the Medium Resolution Spectral Imager (MERSI) on FengYun-3. However, panoramic distortion in whiskbroom scanner images increases overlap from the nadir to the edges between adjacent scans. These distortions present significant challenges for improving geolocation accuracy, particularly when errors occur in sensors and platforms. This manuscript derives analytic expressions for all potential error sources, including sensors, platforms, and elevation, using homogeneous coordinates in the focal plane. This derivation demonstrates that geolocation errors vary with view angles and detector positions. To further investigate these error properties, a gradient-aware least-squares matching method was developed to extract highly accurate and dense ground control points (GCPs) with approximately 100,000 points in a single scene. A three-step geometric calibration method was then introduced, which includes boresight misalignment correction, parametric geometric calibration, and non-uniform scanning compensation. Given the varying spatial resolution of the GCPs, the weight of the GCPs was dynamically updated for least-squares estimation. This method effectively demonstrated the complex geolocation errors in MERSI on FY-3D, a system that was not meticulously calibrated in the laboratory. The initial root mean square errors (RMSEs) were 3.354 and 12.441 instantaneous field of view (IFoV) for the designed parameters. The proposed geometric calibration method successfully corrected view-angle and detector position-related geolocation errors, reducing them to 0.211 and 0.225 IFoV in the scan and track directions, respectively. The geolocation validation software and experiment results were provided <span><span>https://github.com/hongbop/whiskgeovalidation.git</span><svg><path></path></svg></span>.</div></div>","PeriodicalId":50269,"journal":{"name":"ISPRS Journal of Photogrammetry and Remote Sensing","volume":"218 ","pages":"Pages 481-497"},"PeriodicalIF":10.6,"publicationDate":"2024-11-18","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142691122","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"地球科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

ACMatch: Improving context capture for two-view correspondence learning via adaptive convolution ACMatch：通过自适应卷积改进双视角对应学习的上下文捕捉

IF 10.6 1区地球科学 Q1 GEOGRAPHY, PHYSICAL

ISPRS Journal of Photogrammetry and Remote Sensing

Pub Date : 2024-11-16 DOI: 10.1016/j.isprsjprs.2024.11.004

Xiang Fang , Yifan Lu , Shihua Zhang , Yining Xie , Jiayi Ma

Two-view correspondence learning plays a pivotal role in the field of computer vision. However, this task is beset with great challenges stemming from the significant imbalance between true and false correspondences. Recent approaches have started leveraging the inherent filtering properties of convolution to eliminate false matches. Nevertheless, these methods tend to apply convolution in an ad hoc manner without careful design, thereby inheriting the limitations of convolution and hindering performance improvement. In this paper, we propose a novel convolution-based method called ACMatch, which aims to meticulously design convolutional filters to mitigate the shortcomings of convolution and enhance its effectiveness. Specifically, to address the limitation of existing convolutional filters of struggling to effectively capture global information due to the limited receptive field, we introduce a strategy to help them obtain relatively global information by guiding grid points to incorporate more contextual information, thus enabling a global perspective for two-view learning. Furthermore, we recognize that in the context of feature matching, inliers and outliers provide fundamentally different information. Hence, we design an adaptive weighted convolution module that allows the filters to focus more on inliers while ignoring outliers. Extensive experiments across various visual tasks demonstrate the effectiveness, superiority, and generalization. Notably, ACMatch attains an AUC@5° of 35.93% on YFCC100M without RANSAC, surpassing the previous state-of-the-art by 5.85 absolute percentage points and exceeding the 35% AUC@5° bar for the first time. Our code is publicly available at https://github.com/ShineFox/ACMatch.

双视角对应学习在计算机视觉领域发挥着举足轻重的作用。然而，由于真假对应关系严重失衡，这项任务面临着巨大的挑战。最近的方法开始利用卷积的固有过滤特性来消除错误匹配。然而，这些方法往往未经精心设计就临时应用卷积，从而继承了卷积的局限性，阻碍了性能的提高。在本文中，我们提出了一种名为 ACMatch 的基于卷积的新方法，旨在精心设计卷积滤波器，以减轻卷积的缺点并提高其有效性。具体来说，针对现有卷积滤波器因感受野有限而难以有效捕捉全局信息的局限，我们引入了一种策略，通过引导网格点纳入更多上下文信息，帮助它们获取相对全局的信息，从而实现双视角学习的全局视角。此外，我们还认识到，在特征匹配中，异常值和离群值提供了根本不同的信息。因此，我们设计了一个自适应加权卷积模块，允许滤波器更多地关注异常值，而忽略离群值。在各种视觉任务中进行的大量实验证明了 ACMatch 的有效性、优越性和通用性。值得注意的是，在不使用 RANSAC 的情况下，ACMatch 在 YFCC100M 上的 AUC@5° 达到了 35.93%，比之前的一流水平高出 5.85 个绝对百分点，并首次超过了 35% AUC@5° 的标准。我们的代码可在 https://github.com/ShineFox/ACMatch 公开获取。

{"title":"ACMatch: Improving context capture for two-view correspondence learning via adaptive convolution","authors":"Xiang Fang , Yifan Lu , Shihua Zhang , Yining Xie , Jiayi Ma","doi":"10.1016/j.isprsjprs.2024.11.004","DOIUrl":"10.1016/j.isprsjprs.2024.11.004","url":null,"abstract":"<div><div>Two-view correspondence learning plays a pivotal role in the field of computer vision. However, this task is beset with great challenges stemming from the significant imbalance between true and false correspondences. Recent approaches have started leveraging the inherent filtering properties of convolution to eliminate false matches. Nevertheless, these methods tend to apply convolution in an ad hoc manner without careful design, thereby inheriting the limitations of convolution and hindering performance improvement. In this paper, we propose a novel convolution-based method called ACMatch, which aims to meticulously design convolutional filters to mitigate the shortcomings of convolution and enhance its effectiveness. Specifically, to address the limitation of existing convolutional filters of struggling to effectively capture global information due to the limited receptive field, we introduce a strategy to help them obtain relatively global information by guiding grid points to incorporate more contextual information, thus enabling a global perspective for two-view learning. Furthermore, we recognize that in the context of feature matching, inliers and outliers provide fundamentally different information. Hence, we design an adaptive weighted convolution module that allows the filters to focus more on inliers while ignoring outliers. Extensive experiments across various visual tasks demonstrate the effectiveness, superiority, and generalization. Notably, ACMatch attains an AUC@5° of 35.93% on YFCC100M without RANSAC, surpassing the previous state-of-the-art by 5.85 absolute percentage points and exceeding the 35% AUC@5° bar for the first time. Our code is publicly available at <span><span>https://github.com/ShineFox/ACMatch</span><svg><path></path></svg></span>.</div></div>","PeriodicalId":50269,"journal":{"name":"ISPRS Journal of Photogrammetry and Remote Sensing","volume":"218 ","pages":"Pages 466-480"},"PeriodicalIF":10.6,"publicationDate":"2024-11-16","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142658279","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"地球科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

A universal adapter in segmentation models for transferable landslide mapping 用于可转移滑坡绘图的分段模型中的通用适配器

IF 10.6 1区地球科学 Q1 GEOGRAPHY, PHYSICAL

ISPRS Journal of Photogrammetry and Remote Sensing

Pub Date : 2024-11-15 DOI: 10.1016/j.isprsjprs.2024.11.006

Ruilong Wei , Yamei Li , Yao Li , Bo Zhang , Jiao Wang , Chunhao Wu , Shunyu Yao , Chengming Ye

Efficient landslide mapping is crucial for disaster mitigation and relief. Recently, deep learning methods have shown promising results in landslide mapping using satellite imagery. However, the sample sparsity and geographic diversity of landslides have challenged the transferability of deep learning models. In this paper, we proposed a universal adapter module that can be seamlessly embedded into existing segmentation models for transferable landslide mapping. The adapter can achieve high-accuracy cross-regional landslide segmentation with a small sample set, requiring minimal parameter adjustments. In detail, the pre-trained baseline model freezes its parameters to keep learned knowledge of the source domain, while the lightweight adapter fine-tunes only a few parameters to learn new landslide features of the target domain. Structurally, we introduced an attention mechanism to enhance the feature extraction of the adapter. To validate the proposed adapter module, 4321 landslide samples were prepared, and the Segment Anything Model (SAM) and other baseline models, along with four transfer strategies were selected for controlled experiments. In addition, Sentinel-2 satellite imagery in the Himalayas and Hengduan Mountains, located on the southern and southeastern edges of the Tibetan Plateau was collected for evaluation. The controlled experiments reported that SAM, when combined with our adapter module, achieved a peak mean Intersection over Union (mIoU) of 82.3 %. For other baseline models, integrating the adapter improved mIoU by 2.6 % to 12.9 % compared with traditional strategies on cross-regional landslide mapping. In particular, baseline models with Transformers are more suitable for fine-tuning parameters. Furthermore, the visualized feature maps revealed that fine-tuning shallow encoders can achieve better effects in model transfer. Besides, the proposed adapter can effectively extract landslide features and focus on specific spatial and channel domains with significant features. We also quantified the spectral, scale, and shape features of landslides and analyzed their impacts on segmentation results. Our analysis indicated that weak spectral differences, as well as extreme scale and edge shapes are detrimental to the accuracy of landslide segmentation. Overall, this adapter module provides a new perspective for large-scale transferable landslide mapping.

高效的滑坡绘图对减灾救灾至关重要。最近，深度学习方法在利用卫星图像绘制滑坡地图方面取得了可喜的成果。然而，滑坡的样本稀疏性和地理多样性对深度学习模型的可移植性提出了挑战。在本文中，我们提出了一种通用适配器模块，可无缝嵌入现有的细分模型，实现滑坡绘图的可移植性。该适配器只需少量样本集，就能实现高精度的跨区域滑坡分割，参数调整量极小。具体来说，预先训练好的基线模型会冻结其参数，以保留源领域的已学知识，而轻量级适配器只需微调几个参数，就能学习目标领域的新滑坡特征。在结构上，我们引入了注意力机制，以加强适配器的特征提取。为了验证所提出的适配器模块，我们准备了 4321 个滑坡样本，并选择了分段任意模型（SAM）和其他基线模型以及四种转移策略进行对照实验。此外，还收集了位于青藏高原南部和东南部边缘的喜马拉雅山脉和横断山脉的哨兵-2 卫星图像进行评估。对照实验结果表明，当 SAM 与我们的适配器模块相结合时，峰值平均联合交叉率（mIoU）达到了 82.3%。对于其他基线模型，与跨区域滑坡绘图的传统策略相比，集成适配器可将 mIoU 提高 2.6% 至 12.9%。特别是，带有转换器的基线模型更适合微调参数。此外，可视化特征图显示，微调浅层编码器可在模型转移中取得更好的效果。此外，所提出的适配器能有效提取滑坡特征，并聚焦于具有重要特征的特定空间和通道域。我们还量化了滑坡的光谱、尺度和形状特征，并分析了它们对分割结果的影响。我们的分析表明，微弱的光谱差异以及极端的尺度和边缘形状不利于滑坡分割的准确性。总之，该适配器模块为大规模可转移滑坡绘图提供了新的视角。

{"title":"A universal adapter in segmentation models for transferable landslide mapping","authors":"Ruilong Wei , Yamei Li , Yao Li , Bo Zhang , Jiao Wang , Chunhao Wu , Shunyu Yao , Chengming Ye","doi":"10.1016/j.isprsjprs.2024.11.006","DOIUrl":"10.1016/j.isprsjprs.2024.11.006","url":null,"abstract":"<div><div>Efficient landslide mapping is crucial for disaster mitigation and relief. Recently, deep learning methods have shown promising results in landslide mapping using satellite imagery. However, the sample sparsity and geographic diversity of landslides have challenged the transferability of deep learning models. In this paper, we proposed a universal adapter module that can be seamlessly embedded into existing segmentation models for transferable landslide mapping. The adapter can achieve high-accuracy cross-regional landslide segmentation with a small sample set, requiring minimal parameter adjustments. In detail, the pre-trained baseline model freezes its parameters to keep learned knowledge of the source domain, while the lightweight adapter fine-tunes only a few parameters to learn new landslide features of the target domain. Structurally, we introduced an attention mechanism to enhance the feature extraction of the adapter. To validate the proposed adapter module, 4321 landslide samples were prepared, and the Segment Anything Model (SAM) and other baseline models, along with four transfer strategies were selected for controlled experiments. In addition, Sentinel-2 satellite imagery in the Himalayas and Hengduan Mountains, located on the southern and southeastern edges of the Tibetan Plateau was collected for evaluation. The controlled experiments reported that SAM, when combined with our adapter module, achieved a peak mean Intersection over Union (mIoU) of 82.3 %. For other baseline models, integrating the adapter improved mIoU by 2.6 % to 12.9 % compared with traditional strategies on cross-regional landslide mapping. In particular, baseline models with Transformers are more suitable for fine-tuning parameters. Furthermore, the visualized feature maps revealed that fine-tuning shallow encoders can achieve better effects in model transfer. Besides, the proposed adapter can effectively extract landslide features and focus on specific spatial and channel domains with significant features. We also quantified the spectral, scale, and shape features of landslides and analyzed their impacts on segmentation results. Our analysis indicated that weak spectral differences, as well as extreme scale and edge shapes are detrimental to the accuracy of landslide segmentation. Overall, this adapter module provides a new perspective for large-scale transferable landslide mapping.</div></div>","PeriodicalId":50269,"journal":{"name":"ISPRS Journal of Photogrammetry and Remote Sensing","volume":"218 ","pages":"Pages 446-465"},"PeriodicalIF":10.6,"publicationDate":"2024-11-15","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142658278","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"地球科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

Contrastive learning for real SAR image despeckling 针对真实合成孔径雷达图像去斑的对比学习

IF 10.6 1区地球科学 Q1 GEOGRAPHY, PHYSICAL

ISPRS Journal of Photogrammetry and Remote Sensing

Pub Date : 2024-11-15 DOI: 10.1016/j.isprsjprs.2024.11.003

Yangtian Fang , Rui Liu , Yini Peng , Jianjun Guan , Duidui Li , Xin Tian

The use of synthetic aperture radar (SAR) has greatly improved our ability to capture high-resolution terrestrial images under various weather conditions. However, SAR imagery is affected by speckle noise, which distorts image details and hampers subsequent applications. Recent forays into supervised deep learning-based denoising methods, like MRDDANet and SAR-CAM, offer a promising avenue for SAR despeckling. However, they are impeded by the domain gaps between synthetic data and realistic SAR images. To tackle this problem, we introduce a self-supervised speckle-aware network to utilize the limited near-real datasets and unlimited synthetic datasets simultaneously, which boosts the performance of the downstream despeckling module by teaching the module to discriminate the domain gap of different datasets in the embedding space. Specifically, based on contrastive learning, the speckle-aware network first characterizes the discriminative representations of spatial-correlated speckle noise in different images across diverse datasets, which provides priors of versatile speckles and image characteristics. Then, the representations are effectively modulated into a subsequent multi-scale despeckling network to generate authentic despeckled images. In this way, the despeckling module can reconstruct reliable SAR image characteristics by learning from near-real datasets, while the generalization performance is guaranteed by learning abundant patterns from synthetic datasets simultaneously. Additionally, a novel excitation aggregation pooling module is inserted into the despeckling network to enhance the network further, which utilizes features from different levels of scales and better preserves spatial details around strong scatters in real SAR images. Extensive experiments across real SAR datasets from Sentinel-1, Capella-X, and TerraSAR-X satellites are carried out to verify the effectiveness of the proposed method over other state-of-the-art methods. Specifically, the proposed method achieves the best PSNR and SSIM values evaluated on the near-real Sentinel-1 dataset, with gains of 0.22 dB in PSNR compared to MRDDANet, and improvements of 1.3% in SSIM over SAR-CAM. The code is available at https://github.com/YangtianFang2002/CL-SAR-Despeckling.

合成孔径雷达（SAR）的使用大大提高了我们在各种天气条件下捕捉高分辨率陆地图像的能力。然而，合成孔径雷达图像会受到斑点噪声的影响，从而扭曲图像细节，妨碍后续应用。最近，基于监督深度学习的去噪方法（如 MRDDANet 和 SAR-CAM）为合成孔径雷达去斑提供了一条前景广阔的途径。然而，合成数据与现实合成孔径雷达图像之间的领域差距阻碍了它们的发展。为了解决这个问题，我们引入了一种自监督斑点感知网络，同时利用有限的近真实数据集和无限的合成数据集，通过教会模块辨别嵌入空间中不同数据集的域差距，提高下游解斑模块的性能。具体来说，基于对比学习，斑点感知网络首先描述了不同数据集中不同图像中空间相关斑点噪声的判别表征，从而提供了多功能斑点和图像特征的先验。然后，将这些表征有效地调制到随后的多尺度去斑网络中，生成真实的去斑图像。这样，去斑模块就能通过学习近乎真实的数据集来重建可靠的合成孔径雷达图像特征，同时通过同时学习合成数据集的丰富模式来保证泛化性能。此外，除斑网络中还加入了一个新颖的激励聚合池化模块，以进一步增强网络，从而利用不同尺度的特征，更好地保留真实合成孔径雷达图像中强散射周围的空间细节。通过对来自 Sentinel-1、Capella-X 和 TerraSAR-X 卫星的真实合成孔径雷达数据集进行广泛实验，验证了所提方法相对于其他先进方法的有效性。具体来说，在近乎真实的 Sentinel-1 数据集上，所提方法获得了最佳的 PSNR 和 SSIM 值，PSNR 比 MRDDANet 提高了 0.22 dB，SSIM 比 SAR-CAM 提高了 1.3%。代码见 https://github.com/YangtianFang2002/CL-SAR-Despeckling。

{"title":"Contrastive learning for real SAR image despeckling","authors":"Yangtian Fang , Rui Liu , Yini Peng , Jianjun Guan , Duidui Li , Xin Tian","doi":"10.1016/j.isprsjprs.2024.11.003","DOIUrl":"10.1016/j.isprsjprs.2024.11.003","url":null,"abstract":"<div><div>The use of synthetic aperture radar (SAR) has greatly improved our ability to capture high-resolution terrestrial images under various weather conditions. However, SAR imagery is affected by speckle noise, which distorts image details and hampers subsequent applications. Recent forays into supervised deep learning-based denoising methods, like MRDDANet and SAR-CAM, offer a promising avenue for SAR despeckling. However, they are impeded by the domain gaps between synthetic data and realistic SAR images. To tackle this problem, we introduce a self-supervised speckle-aware network to utilize the limited near-real datasets and unlimited synthetic datasets simultaneously, which boosts the performance of the downstream despeckling module by teaching the module to discriminate the domain gap of different datasets in the embedding space. Specifically, based on contrastive learning, the speckle-aware network first characterizes the discriminative representations of spatial-correlated speckle noise in different images across diverse datasets, which provides priors of versatile speckles and image characteristics. Then, the representations are effectively modulated into a subsequent multi-scale despeckling network to generate authentic despeckled images. In this way, the despeckling module can reconstruct reliable SAR image characteristics by learning from near-real datasets, while the generalization performance is guaranteed by learning abundant patterns from synthetic datasets simultaneously. Additionally, a novel excitation aggregation pooling module is inserted into the despeckling network to enhance the network further, which utilizes features from different levels of scales and better preserves spatial details around strong scatters in real SAR images. Extensive experiments across real SAR datasets from Sentinel-1, Capella-X, and TerraSAR-X satellites are carried out to verify the effectiveness of the proposed method over other state-of-the-art methods. Specifically, the proposed method achieves the best PSNR and SSIM values evaluated on the near-real Sentinel-1 dataset, with gains of 0.22 dB in PSNR compared to MRDDANet, and improvements of 1.3% in SSIM over SAR-CAM. The code is available at <span><span>https://github.com/YangtianFang2002/CL-SAR-Despeckling</span><svg><path></path></svg></span>.</div></div>","PeriodicalId":50269,"journal":{"name":"ISPRS Journal of Photogrammetry and Remote Sensing","volume":"218 ","pages":"Pages 376-391"},"PeriodicalIF":10.6,"publicationDate":"2024-11-15","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142658338","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"地球科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

MIWC: A multi-temporal image weighted composition method for satellite-derived bathymetry in shallow waters MIWC：用于浅水卫星水深测量的多时相图像加权合成法

IF 10.6 1区地球科学 Q1 GEOGRAPHY, PHYSICAL

ISPRS Journal of Photogrammetry and Remote Sensing

Pub Date : 2024-11-15 DOI: 10.1016/j.isprsjprs.2024.10.009

Zhixin Duan , Liang Cheng , Qingzhou Mao , Yueting Song , Xiao Zhou , Manchun Li , Jianya Gong

Satellite-derived bathymetry (SDB) is a vital technique for the rapid and cost-effective measurement of shallow underwater terrain. However, it faces challenges of image noise, including clouds, bubble clouds, and sun glint. Consequently, the acquisition of no missing and accurate bathymetric maps is frequently challenging, particularly in cloudy, rainy, and large-scale regions. In this study, we propose a multi-temporal image weighted composition (MIWC) method. This method performs iterative segmentation and inverse distance weighted composition of multi-temporal images based only on the near-infrared (NIR) band information of multispectral images to obtain high-quality composite images. The method was applied to scenarios using Sentinel-2 imagery for bathymetry of four representative areas located in the South China Sea and the Atlantic Ocean. The results show that the root mean square error (RMSE) of bathymetry from the composite images using the log-transformed linear model (LLM) and the log-transformed ratio model (LRM) in the water depth range of 0–20 m are 0.67–1.22 m and 0.71–1.23 m, respectively. The RMSE of the bathymetry decreases with the number of images involved in the composition and tends to be relatively stable when the number of images reaches approximately 16. In addition, the composition images generated by the MIWC method generally exhibit not only superior visual quality, but also significant advantages in terms of bathymetric accuracy and robustness when compared to the best single images as well as the composition images generated by the median composition method and the maximum outlier removal method. The recommended value of the power parameter for inverse distance weighting in the MIWC method was experimentally determined to be 4, which typically does not require complex adjustments, making the method easy to apply or integrate. The MIWC method offers a reliable approach to improve the quality of remote sensing images, ensuring the completeness and accuracy of shallow water bathymetric maps.

卫星水深测量（SDB）是快速、经济地测量水下浅层地形的重要技术。然而，它面临着图像噪声的挑战，包括云层、气泡云和太阳光。因此，特别是在多云、多雨和大尺度地区，获取无遗漏和准确的水深测量图经常面临挑战。在本研究中，我们提出了一种多时相图像加权合成（MIWC）方法。该方法仅基于多光谱图像的近红外（NIR）波段信息，对多时态图像进行迭代分割和反距离加权合成，从而获得高质量的合成图像。该方法被应用于使用哨兵-2 图像对位于中国南海和大西洋的四个代表性区域进行水深测量的场景。结果表明，在 0-20 米水深范围内，使用对数变换线性模型（LLM）和对数变换比值模型（LRM）从合成图像得出的水深测量结果的均方根误差（RMSE）分别为 0.67-1.22 米和 0.71-1.23 米。水深测量的均方根误差随组成图像数量的增加而减小，当图像数量达到约 16 幅时，均方根误差趋于相对稳定。此外，与最佳单幅图像以及中值合成法和最大离群点去除法生成的合成图像相比，MIWC 方法生成的合成图像不仅视觉质量上乘，而且在测深精度和鲁棒性方面也有显著优势。经实验确定，MIWC 方法中用于反距离加权的功率参数推荐值为 4，通常不需要进行复杂的调整，因此该方法易于应用或集成。MIWC 方法是提高遥感图像质量的可靠方法，可确保浅水测深图的完整性和准确性。

{"title":"MIWC: A multi-temporal image weighted composition method for satellite-derived bathymetry in shallow waters","authors":"Zhixin Duan , Liang Cheng , Qingzhou Mao , Yueting Song , Xiao Zhou , Manchun Li , Jianya Gong","doi":"10.1016/j.isprsjprs.2024.10.009","DOIUrl":"10.1016/j.isprsjprs.2024.10.009","url":null,"abstract":"<div><div>Satellite-derived bathymetry (SDB) is a vital technique for the rapid and cost-effective measurement of shallow underwater terrain. However, it faces challenges of image noise, including clouds, bubble clouds, and sun glint. Consequently, the acquisition of no missing and accurate bathymetric maps is frequently challenging, particularly in cloudy, rainy, and large-scale regions. In this study, we propose a multi-temporal image weighted composition (MIWC) method. This method performs iterative segmentation and inverse distance weighted composition of multi-temporal images based only on the near-infrared (NIR) band information of multispectral images to obtain high-quality composite images. The method was applied to scenarios using Sentinel-2 imagery for bathymetry of four representative areas located in the South China Sea and the Atlantic Ocean. The results show that the root mean square error (RMSE) of bathymetry from the composite images using the log-transformed linear model (LLM) and the log-transformed ratio model (LRM) in the water depth range of 0–20 m are 0.67–1.22 m and 0.71–1.23 m, respectively. The RMSE of the bathymetry decreases with the number of images involved in the composition and tends to be relatively stable when the number of images reaches approximately 16. In addition, the composition images generated by the MIWC method generally exhibit not only superior visual quality, but also significant advantages in terms of bathymetric accuracy and robustness when compared to the best single images as well as the composition images generated by the median composition method and the maximum outlier removal method. The recommended value of the power parameter for inverse distance weighting in the MIWC method was experimentally determined to be 4, which typically does not require complex adjustments, making the method easy to apply or integrate. The MIWC method offers a reliable approach to improve the quality of remote sensing images, ensuring the completeness and accuracy of shallow water bathymetric maps.</div></div>","PeriodicalId":50269,"journal":{"name":"ISPRS Journal of Photogrammetry and Remote Sensing","volume":"218 ","pages":"Pages 430-445"},"PeriodicalIF":10.6,"publicationDate":"2024-11-15","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142658277","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"地球科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

Common-feature-track-matching approach for multi-epoch UAV photogrammetry co-registration 多波段无人机摄影测量共准法的共同特征轨迹匹配方法

IF 10.6 1区地球科学 Q1 GEOGRAPHY, PHYSICAL

ISPRS Journal of Photogrammetry and Remote Sensing

Pub Date : 2024-11-14 DOI: 10.1016/j.isprsjprs.2024.10.025

Xinlong Li , Mingtao Ding , Zhenhong Li , Peng Cui

Automatic co-registration of multi-epoch Unmanned Aerial Vehicle (UAV) image sets remains challenging due to the radiometric differences in complex dynamic scenes. Specifically, illumination changes and vegetation variations usually lead to insufficient and spatially unevenly distributed common tie points (CTPs), resulting in under-fitting of co-registration near the areas without CTPs. In this paper, we propose a novel Common-Feature-Track-Matching (CFTM) approach for UAV image sets co-registration, to alleviate the shortage of CTPs in complex dynamic scenes. Instead of matching features between multi-epoch images, we first search correspondences between multi-epoch feature tracks (i.e., groups of features corresponding to the same 3D points), which avoids the removal of matches due to unreliable estimation of the relative pose between inter-epoch image pairs. Then, the CTPs are triangulated from the successfully matched track pairs. Since an even distribution of CTPs is crucial for robust co-registration, a block-based strategy is designed, as well as enabling parallel computation. Finally, an iterative optimization algorithm is developed to gradually select the best CTPs to refine the poses of multi-epoch images. We assess the performance of our method on two challenging datasets. The results show that CFTM can automatically acquire adequate and evenly distributed CTPs in complex dynamic scenes, achieving a high co-registration accuracy approximately four times higher than the state-of-the-art in challenging scenario. Our code is available at https://github.com/lixinlong1998/CoSfM.

由于复杂动态场景中的辐射测量差异，多波长无人机（UAV）图像集的自动共配准仍然具有挑战性。具体来说，光照变化和植被变化通常会导致公共连接点（CTP）不足且在空间上分布不均，从而导致在没有公共连接点的区域附近的协同注册拟合不足。在本文中，我们提出了一种用于无人机图像集协同注册的新型公共特征轨迹匹配（CFTM）方法，以缓解复杂动态场景中公共连接点不足的问题。我们首先搜索多时序特征轨迹（即对应于相同三维点的特征组）之间的对应关系，而不是匹配多时序图像之间的特征，这避免了由于对时序间图像对的相对姿态估计不可靠而删除匹配。然后，根据成功匹配的轨迹对进行 CTP 三角测量。由于 CTPs 的均匀分布对稳健的协同注册至关重要，因此设计了一种基于块的策略，并实现了并行计算。最后，我们开发了一种迭代优化算法，以逐步选择最佳 CTP，从而完善多波段图像的姿态。我们在两个具有挑战性的数据集上评估了我们方法的性能。结果表明，CFTM 可以在复杂的动态场景中自动获取足够且分布均匀的 CTP，在具有挑战性的场景中实现了比最先进方法高约四倍的高协同注册精度。我们的代码见 https://github.com/lixinlong1998/CoSfM。

{"title":"Common-feature-track-matching approach for multi-epoch UAV photogrammetry co-registration","authors":"Xinlong Li , Mingtao Ding , Zhenhong Li , Peng Cui","doi":"10.1016/j.isprsjprs.2024.10.025","DOIUrl":"10.1016/j.isprsjprs.2024.10.025","url":null,"abstract":"<div><div>Automatic co-registration of multi-epoch Unmanned Aerial Vehicle (UAV) image sets remains challenging due to the radiometric differences in complex dynamic scenes. Specifically, illumination changes and vegetation variations usually lead to insufficient and spatially unevenly distributed common tie points (CTPs), resulting in under-fitting of co-registration near the areas without CTPs. In this paper, we propose a novel Common-Feature-Track-Matching (CFTM) approach for UAV image sets co-registration, to alleviate the shortage of CTPs in complex dynamic scenes. Instead of matching features between multi-epoch images, we first search correspondences between multi-epoch feature tracks (i.e., groups of features corresponding to the same 3D points), which avoids the removal of matches due to unreliable estimation of the relative pose between inter-epoch image pairs. Then, the CTPs are triangulated from the successfully matched track pairs. Since an even distribution of CTPs is crucial for robust co-registration, a block-based strategy is designed, as well as enabling parallel computation. Finally, an iterative optimization algorithm is developed to gradually select the best CTPs to refine the poses of multi-epoch images. We assess the performance of our method on two challenging datasets. The results show that CFTM can automatically acquire adequate and evenly distributed CTPs in complex dynamic scenes, achieving a high co-registration accuracy approximately four times higher than the state-of-the-art in challenging scenario. Our code is available at <span><span>https://github.com/lixinlong1998/CoSfM</span><svg><path></path></svg></span>.</div></div>","PeriodicalId":50269,"journal":{"name":"ISPRS Journal of Photogrammetry and Remote Sensing","volume":"218 ","pages":"Pages 392-407"},"PeriodicalIF":10.6,"publicationDate":"2024-11-14","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142658339","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"地球科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0