Pub Date : 2025-03-14DOI: 10.1109/LGRS.2025.3551235
Jianfei Chen;Jiahao Yu;Yujie Ruan;Chenggong Zhang;Ziang Zheng;Fuxin Cai;Shujin Zhu;Leilei Liu
The Synthetic Aperture Interferometric Radiometer (SAIR) can realize high-resolution real-time imaging observation by using aperture synthesis technology, which has strong application advantages in the field of earth remote sensing and radio astronomy. However, due to the limitation of engineering technology, the aperture of the SAIR is still limited, which limits the further improvement of SAIR’s spatial resolution. Therefore, this letter proposes a novel super-resolution imaging method based on spectral extrapolation network (SR-SEN), which can further improve the SAIR’s imaging resolution without increasing the system hardware scale. In the SR-SEN method, the spectral extrapolation subnet is used to deduce the high-frequency spectral components from the low-frequency visibility function measured by the SAIR system, and the iterative reconstruction subnet is constructed to realize the super-resolution imaging inversion of the target scene. The simulation results show that the proposed SR-SEN method can realize accurate spectral extrapolation, improve the imaging resolution of SAIR, and finally realize high-quality imaging inversion.
{"title":"Super-Resolution Imaging Method for Synthetic Aperture Interferometric Radiometer Based on Spectral Extrapolation","authors":"Jianfei Chen;Jiahao Yu;Yujie Ruan;Chenggong Zhang;Ziang Zheng;Fuxin Cai;Shujin Zhu;Leilei Liu","doi":"10.1109/LGRS.2025.3551235","DOIUrl":"https://doi.org/10.1109/LGRS.2025.3551235","url":null,"abstract":"The Synthetic Aperture Interferometric Radiometer (SAIR) can realize high-resolution real-time imaging observation by using aperture synthesis technology, which has strong application advantages in the field of earth remote sensing and radio astronomy. However, due to the limitation of engineering technology, the aperture of the SAIR is still limited, which limits the further improvement of SAIR’s spatial resolution. Therefore, this letter proposes a novel super-resolution imaging method based on spectral extrapolation network (SR-SEN), which can further improve the SAIR’s imaging resolution without increasing the system hardware scale. In the SR-SEN method, the spectral extrapolation subnet is used to deduce the high-frequency spectral components from the low-frequency visibility function measured by the SAIR system, and the iterative reconstruction subnet is constructed to realize the super-resolution imaging inversion of the target scene. The simulation results show that the proposed SR-SEN method can realize accurate spectral extrapolation, improve the imaging resolution of SAIR, and finally realize high-quality imaging inversion.","PeriodicalId":91017,"journal":{"name":"IEEE geoscience and remote sensing letters : a publication of the IEEE Geoscience and Remote Sensing Society","volume":"22 ","pages":"1-5"},"PeriodicalIF":0.0,"publicationDate":"2025-03-14","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143716479","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2025-03-13DOI: 10.1109/LGRS.2025.3550967
Juan Pablo Navarro Castillo;Rolf Scheiber;Marc Jäger;Alberto Moreira
High-resolution wide swath (HRWS) synthetic aperture radar (SAR) systems are normally designed to have identical antenna patterns in each receive channel. Nevertheless, due to different factors, this condition might not be satisfied, resulting in a multichannel radar system with different antenna patterns. This letter studies the impact of these relative antenna differences on the performance of a state-of-the-art azimuth motion-adaptive image reconstruction for a real airborne SAR sensor with multiple azimuth channels on receive. The performance of the reconstruction algorithm is evaluated both when these relative differences are neglected and when they are accounted for in the multichannel reconstruction process. Additionally, the range dependence of the antenna pattern as a function of wavenumber and the use of range block processing to minimize the impact of this dependence are analyzed. The results confirm the relevance of including these additional steps in the reconstruction, extending the understanding of SAR systems using digital beamforming (DBF) in azimuth.
{"title":"Interchannel Antenna Pattern Correction for Azimuth Digital Beamforming in Airborne SAR","authors":"Juan Pablo Navarro Castillo;Rolf Scheiber;Marc Jäger;Alberto Moreira","doi":"10.1109/LGRS.2025.3550967","DOIUrl":"https://doi.org/10.1109/LGRS.2025.3550967","url":null,"abstract":"High-resolution wide swath (HRWS) synthetic aperture radar (SAR) systems are normally designed to have identical antenna patterns in each receive channel. Nevertheless, due to different factors, this condition might not be satisfied, resulting in a multichannel radar system with different antenna patterns. This letter studies the impact of these relative antenna differences on the performance of a state-of-the-art azimuth motion-adaptive image reconstruction for a real airborne SAR sensor with multiple azimuth channels on receive. The performance of the reconstruction algorithm is evaluated both when these relative differences are neglected and when they are accounted for in the multichannel reconstruction process. Additionally, the range dependence of the antenna pattern as a function of wavenumber and the use of range block processing to minimize the impact of this dependence are analyzed. The results confirm the relevance of including these additional steps in the reconstruction, extending the understanding of SAR systems using digital beamforming (DBF) in azimuth.","PeriodicalId":91017,"journal":{"name":"IEEE geoscience and remote sensing letters : a publication of the IEEE Geoscience and Remote Sensing Society","volume":"22 ","pages":"1-5"},"PeriodicalIF":0.0,"publicationDate":"2025-03-13","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143716470","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
In this letter, an Inverse synthetic aperture radar (ISAR) imaging method uses second harmonic signals to extract motion parameters of multi-rotor autonomous aerial vehicles in an environment containing other natural objects for high-precision ISAR imaging. The core of the drones, which include powerful radio frequency (RF) circuits, is proven to have strong nonlinear effects. Translational motion parameters are estimated from prominent points in second harmonic signals, and the rotational motion of the drone is corrected by instantaneous imaging. An integrated RF system commonly used in drones has been tested, and its nonlinear phase characteristics have been obtained. The real phase error due to the nonlinearity was carried over into the subsequent simulation and compensated, which did not lead to defocusing in the final image. The feasibility of using second harmonic signals to compensate for fundamental frequency motion errors is verified by simulations. The micro-Doppler effect, generated by the high-speed rotation of the blades, has been demonstrated in the simulation images. System integration with a harmonic signal-based motion correction (H-MC) approach for motion compensation is shown to demonstrate a solution for accurate imaging for drones.
{"title":"High-Precision ISAR Method for Nonlinear Rotating Targets: Imaging for Drones","authors":"Chenhao Zhao;Qinghai Dong;Bingnan Wang;Maosheng Xiang","doi":"10.1109/LGRS.2025.3547441","DOIUrl":"https://doi.org/10.1109/LGRS.2025.3547441","url":null,"abstract":"In this letter, an Inverse synthetic aperture radar (ISAR) imaging method uses second harmonic signals to extract motion parameters of multi-rotor autonomous aerial vehicles in an environment containing other natural objects for high-precision ISAR imaging. The core of the drones, which include powerful radio frequency (RF) circuits, is proven to have strong nonlinear effects. Translational motion parameters are estimated from prominent points in second harmonic signals, and the rotational motion of the drone is corrected by instantaneous imaging. An integrated RF system commonly used in drones has been tested, and its nonlinear phase characteristics have been obtained. The real phase error due to the nonlinearity was carried over into the subsequent simulation and compensated, which did not lead to defocusing in the final image. The feasibility of using second harmonic signals to compensate for fundamental frequency motion errors is verified by simulations. The micro-Doppler effect, generated by the high-speed rotation of the blades, has been demonstrated in the simulation images. System integration with a harmonic signal-based motion correction (H-MC) approach for motion compensation is shown to demonstrate a solution for accurate imaging for drones.","PeriodicalId":91017,"journal":{"name":"IEEE geoscience and remote sensing letters : a publication of the IEEE Geoscience and Remote Sensing Society","volume":"22 ","pages":"1-5"},"PeriodicalIF":0.0,"publicationDate":"2025-03-12","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143716495","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Accurate lunar rock detection is vital for lunar exploration. However, the existing methods are sensitive to factors, such as the uneven lighting and terrain relief. To address these issues, a novel method combining multiscale phase feature type maps (PFTMs) and phase congruency moment maps (PCMMs) is proposed. First, rock seeds are detected through phase congruency and gradient analysis. Second, a strategy called the local scale saliency score (LSSS) is proposed to adaptively estimate the optimal scale layer for candidate rock detection. Within this layer, the specifically designed local-contextual and global-contextual (LGC) features are employed to identify the regions of interests (ROIs) for rocks. Subsequently, the process of filtering false positives (FPs) involves the utilization of geometric metrics and scale feature analysis. Finally, a specially designed edge detector named the bilateral local maximum PCMM-PFTM path is proposed to describe the edges of the rocks. Tests on Chang’E-3 and Chang’E-5 Landing Camera (LCAM) images show the proposed method’s robustness in detecting lunar rocks of varying sizes and reflectance, achieving F1-scores ranging from 0.919 to 0.947.
{"title":"A Novel Lunar Rock Detection Method Combining Multiscale Phase Feature Type Maps and Phase Congruency Moment Maps","authors":"Yaqiong Wang;Huan Xie;Qian Huang;Yifan Wang;Xiongfeng Yan;Xiaohua Tong;Shijie Liu;Zhen Ye;Sicong Liu;Xiong Xu;Chao Wang","doi":"10.1109/LGRS.2025.3550461","DOIUrl":"https://doi.org/10.1109/LGRS.2025.3550461","url":null,"abstract":"Accurate lunar rock detection is vital for lunar exploration. However, the existing methods are sensitive to factors, such as the uneven lighting and terrain relief. To address these issues, a novel method combining multiscale phase feature type maps (PFTMs) and phase congruency moment maps (PCMMs) is proposed. First, rock seeds are detected through phase congruency and gradient analysis. Second, a strategy called the local scale saliency score (LSSS) is proposed to adaptively estimate the optimal scale layer for candidate rock detection. Within this layer, the specifically designed local-contextual and global-contextual (LGC) features are employed to identify the regions of interests (ROIs) for rocks. Subsequently, the process of filtering false positives (FPs) involves the utilization of geometric metrics and scale feature analysis. Finally, a specially designed edge detector named the bilateral local maximum PCMM-PFTM path is proposed to describe the edges of the rocks. Tests on Chang’E-3 and Chang’E-5 Landing Camera (LCAM) images show the proposed method’s robustness in detecting lunar rocks of varying sizes and reflectance, achieving F1-scores ranging from 0.919 to 0.947.","PeriodicalId":91017,"journal":{"name":"IEEE geoscience and remote sensing letters : a publication of the IEEE Geoscience and Remote Sensing Society","volume":"22 ","pages":"1-5"},"PeriodicalIF":0.0,"publicationDate":"2025-03-12","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143735405","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2025-03-12DOI: 10.1109/LGRS.2025.3550682
Ali Gholami;Silvia Gazzola
Full waveform inversion (FWI) is a challenging, ill-posed nonlinear inverse problem that requires robust regularization techniques to stabilize the solution and yield geologically meaningful results, especially when dealing with sparse data. Standard Tikhonov regularization, though commonly used in FWI, applies uniform smoothing that often leads to oversmoothing of key geological features, as it fails to account for the underlying structural complexity of the subsurface. To overcome this limitation, we propose an FWI algorithm enhanced by a novel Tikhonov regularization technique involving a parametric regularizer, which is automatically optimized to apply directional space-variant smoothing. Specifically, the parameters defining the regularizer (orientation and anisotropy) are treated as additional unknowns in the objective function, allowing the algorithm to estimate them simultaneously with the model. We introduce an efficient numerical implementation for FWI with the proposed space-variant regularization. Numerical tests on sparse data demonstrate the proposed method’s effectiveness and robustness in reconstructing models with complex structures, significantly improving the inversion results compared with the standard Tikhonov regularization.
{"title":"Optimal Space-Variant Anisotropic Tikhonov Regularization for Full Waveform Inversion of Sparse Data","authors":"Ali Gholami;Silvia Gazzola","doi":"10.1109/LGRS.2025.3550682","DOIUrl":"https://doi.org/10.1109/LGRS.2025.3550682","url":null,"abstract":"Full waveform inversion (FWI) is a challenging, ill-posed nonlinear inverse problem that requires robust regularization techniques to stabilize the solution and yield geologically meaningful results, especially when dealing with sparse data. Standard Tikhonov regularization, though commonly used in FWI, applies uniform smoothing that often leads to oversmoothing of key geological features, as it fails to account for the underlying structural complexity of the subsurface. To overcome this limitation, we propose an FWI algorithm enhanced by a novel Tikhonov regularization technique involving a parametric regularizer, which is automatically optimized to apply directional space-variant smoothing. Specifically, the parameters defining the regularizer (orientation and anisotropy) are treated as additional unknowns in the objective function, allowing the algorithm to estimate them simultaneously with the model. We introduce an efficient numerical implementation for FWI with the proposed space-variant regularization. Numerical tests on sparse data demonstrate the proposed method’s effectiveness and robustness in reconstructing models with complex structures, significantly improving the inversion results compared with the standard Tikhonov regularization.","PeriodicalId":91017,"journal":{"name":"IEEE geoscience and remote sensing letters : a publication of the IEEE Geoscience and Remote Sensing Society","volume":"22 ","pages":"1-5"},"PeriodicalIF":0.0,"publicationDate":"2025-03-12","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=10924238","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143740367","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2025-03-11DOI: 10.1109/LGRS.2025.3550408
Tamar Klein;Tom Aizenberg;Roi Ronen
Climate studies often rely on remotely sensed images to retrieve 2-D maps of cloud properties. To advance volumetric analysis, we focus on recovering the 3-D heterogeneous extinction coefficient field of shallow clouds using multiview remote sensing data. Climate research requires large-scale worldwide statistics. To enable scalable data processing, previous deep neural networks (DNNs) can infer at spaceborne remote sensing downlink rates. However, prior methods are limited to a fixed solar illumination direction. In this work, we introduce the first scalable DNN-based system for 3-D cloud retrieval that accommodates varying camera positions and solar directions. By integrating multiview cloud intensity images with camera position and solar direction data, we achieve greater flexibility in recovery. Training of the DNN is performed by a novel two-stage scheme to address the high number of degrees of freedom in this problem. Our approach shows substantial improvements over previous state-of-the-art methods, particularly in handling variations in the sun’s zenith angle.
{"title":"DNN-Based 3-D Cloud Retrieval for Variable Solar Illumination and Multiview Spaceborne Imaging","authors":"Tamar Klein;Tom Aizenberg;Roi Ronen","doi":"10.1109/LGRS.2025.3550408","DOIUrl":"https://doi.org/10.1109/LGRS.2025.3550408","url":null,"abstract":"Climate studies often rely on remotely sensed images to retrieve 2-D maps of cloud properties. To advance volumetric analysis, we focus on recovering the 3-D heterogeneous extinction coefficient field of shallow clouds using multiview remote sensing data. Climate research requires large-scale worldwide statistics. To enable scalable data processing, previous deep neural networks (DNNs) can infer at spaceborne remote sensing downlink rates. However, prior methods are limited to a fixed solar illumination direction. In this work, we introduce the first scalable DNN-based system for 3-D cloud retrieval that accommodates varying camera positions and solar directions. By integrating multiview cloud intensity images with camera position and solar direction data, we achieve greater flexibility in recovery. Training of the DNN is performed by a novel two-stage scheme to address the high number of degrees of freedom in this problem. Our approach shows substantial improvements over previous state-of-the-art methods, particularly in handling variations in the sun’s zenith angle.","PeriodicalId":91017,"journal":{"name":"IEEE geoscience and remote sensing letters : a publication of the IEEE Geoscience and Remote Sensing Society","volume":"22 ","pages":"1-5"},"PeriodicalIF":0.0,"publicationDate":"2025-03-11","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143716496","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2025-03-11DOI: 10.1109/LGRS.2025.3550409
Chi Zhang;Genwang Liu;Chenghui Cao;Jun Sun;Yongshou Dai;Xi Zhang
Rapid and accurate prediction of the sea clutter amplitude distribution is essential to improve target detection capability in synthetic aperture radar (SAR) imagery. In this letter, we propose a sea clutter amplitude network (SCA-Net) based on multitask learning for sea clutter amplitude distribution prediction (SCADP) of SAR images. To reduce the number of model parameters, we design a shallow residual network structure with four residual blocks and replace the normal convolution with depthwise separable convolution in the residual blocks. The efficient channel attention (ECA) module is incorporated into each residual block to strengthen the model’s feature extraction capability. To validate the performance of the model, we construct a SCADP dataset using GaoFen-3 wave mode data. The experimental results on the SCADP dataset indicate that the proposed method achieves the highest prediction accuracy, which proves that the method can effectively achieve integrated prediction of amplitude distribution types and parameters of sea clutter.
{"title":"SCA-Net: A Network Based on Multitask Learning for Sea Clutter Amplitude Distribution Prediction of SAR Images","authors":"Chi Zhang;Genwang Liu;Chenghui Cao;Jun Sun;Yongshou Dai;Xi Zhang","doi":"10.1109/LGRS.2025.3550409","DOIUrl":"https://doi.org/10.1109/LGRS.2025.3550409","url":null,"abstract":"Rapid and accurate prediction of the sea clutter amplitude distribution is essential to improve target detection capability in synthetic aperture radar (SAR) imagery. In this letter, we propose a sea clutter amplitude network (SCA-Net) based on multitask learning for sea clutter amplitude distribution prediction (SCADP) of SAR images. To reduce the number of model parameters, we design a shallow residual network structure with four residual blocks and replace the normal convolution with depthwise separable convolution in the residual blocks. The efficient channel attention (ECA) module is incorporated into each residual block to strengthen the model’s feature extraction capability. To validate the performance of the model, we construct a SCADP dataset using GaoFen-3 wave mode data. The experimental results on the SCADP dataset indicate that the proposed method achieves the highest prediction accuracy, which proves that the method can effectively achieve integrated prediction of amplitude distribution types and parameters of sea clutter.","PeriodicalId":91017,"journal":{"name":"IEEE geoscience and remote sensing letters : a publication of the IEEE Geoscience and Remote Sensing Society","volume":"22 ","pages":"1-5"},"PeriodicalIF":0.0,"publicationDate":"2025-03-11","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143698327","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
As we all know, phase unwrapping (PhU) is one of the key steps affecting interferometric synthetic aperture radar (InSAR) data processing. However, due to the residues, it is difficult to obtain ideal results in the areas with high noise and large-gradient changes. Therefore, how to effectively deal with residues becomes the top priority of the PhU. To address this issue, in this letter, a novel residue degenerate PhU (RDPhU) method is proposed. We use the fast iterative shrinkage thresholding algorithm (FISTA) to solve the residue degradation problem, which introduces a novel branch-cut strategy that can effectively prevent error propagation. To the best of our knowledge, FISTA is first applied to the PhU residues degradation problem. In addition, we introduce regularization theory into $L^{1}$ -norm PhU to further improve the robustness of PhU. More interestingly, the RDPhU method can effectively solve the problem of low accuracy of PhU in the areas with large-gradient changes, while the PhU efficiency of the RDPhU method is greatly improved. Through simulation and TanDEM-X InSAR datasets, it is proved that the proposed method is an efficient and high-accuracy PhU method.
{"title":"A Novel Residue Degenerate Phase Unwrapping Method Using the L¹-Norm","authors":"YanDong Gao;Chao Yan;Wei Zhou;NanShan Zheng;YaChun Mao;ShiJin Li;BinHe Ji;Hefang Bian","doi":"10.1109/LGRS.2025.3549511","DOIUrl":"https://doi.org/10.1109/LGRS.2025.3549511","url":null,"abstract":"As we all know, phase unwrapping (PhU) is one of the key steps affecting interferometric synthetic aperture radar (InSAR) data processing. However, due to the residues, it is difficult to obtain ideal results in the areas with high noise and large-gradient changes. Therefore, how to effectively deal with residues becomes the top priority of the PhU. To address this issue, in this letter, a novel residue degenerate PhU (RDPhU) method is proposed. We use the fast iterative shrinkage thresholding algorithm (FISTA) to solve the residue degradation problem, which introduces a novel branch-cut strategy that can effectively prevent error propagation. To the best of our knowledge, FISTA is first applied to the PhU residues degradation problem. In addition, we introduce regularization theory into <inline-formula> <tex-math>$L^{1}$ </tex-math></inline-formula>-norm PhU to further improve the robustness of PhU. More interestingly, the RDPhU method can effectively solve the problem of low accuracy of PhU in the areas with large-gradient changes, while the PhU efficiency of the RDPhU method is greatly improved. Through simulation and TanDEM-X InSAR datasets, it is proved that the proposed method is an efficient and high-accuracy PhU method.","PeriodicalId":91017,"journal":{"name":"IEEE geoscience and remote sensing letters : a publication of the IEEE Geoscience and Remote Sensing Society","volume":"22 ","pages":"1-5"},"PeriodicalIF":0.0,"publicationDate":"2025-03-10","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143698345","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Underwater mine-like object (MLO) detection with deep learning suffers from limitations due to the scarcity of real-world side-scan sonar (SSS) data. This scarcity leads to overfitting, where models perform well on training data but poorly on unseen data. In this letter, we propose a synthetic to real (Syn2Real) domain generalization approach using diffusion models to address this challenge. Synthetic data generated by DDPM and DDIM models effectively enhances the training dataset. The residual noise in the final sampled images improves the model’s ability to generalize to real-world data with inherent noise and high variation. The baseline mask-region-based convolutional neural network (RCNN) model when trained on a combination of synthetic and original SSS training datasets, exhibited approximately a 35% increase in average precision (AP) compared to being trained solely on the original training data. This significant improvement highlights the potential of Syn2Real domain generalization for underwater mine detection.
{"title":"Syn2Real Domain Generalization for Underwater Mine-Like Object Detection Using Side-Scan Sonar","authors":"Aayush Agrawal;Aniruddh Sikdar;Rajini Makam;Suresh Sundaram;Suresh Kumar Besai;Mahesh Gopi","doi":"10.1109/LGRS.2025.3550037","DOIUrl":"https://doi.org/10.1109/LGRS.2025.3550037","url":null,"abstract":"Underwater mine-like object (MLO) detection with deep learning suffers from limitations due to the scarcity of real-world side-scan sonar (SSS) data. This scarcity leads to overfitting, where models perform well on training data but poorly on unseen data. In this letter, we propose a synthetic to real (Syn2Real) domain generalization approach using diffusion models to address this challenge. Synthetic data generated by DDPM and DDIM models effectively enhances the training dataset. The residual noise in the final sampled images improves the model’s ability to generalize to real-world data with inherent noise and high variation. The baseline mask-region-based convolutional neural network (RCNN) model when trained on a combination of synthetic and original SSS training datasets, exhibited approximately a 35% increase in average precision (AP) compared to being trained solely on the original training data. This significant improvement highlights the potential of Syn2Real domain generalization for underwater mine detection.","PeriodicalId":91017,"journal":{"name":"IEEE geoscience and remote sensing letters : a publication of the IEEE Geoscience and Remote Sensing Society","volume":"22 ","pages":"1-5"},"PeriodicalIF":0.0,"publicationDate":"2025-03-10","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143716480","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
In the calibration of distributed radar for radio astronomy, deep space radio sources are commonly used as calibration sources to correct interarray delay errors, and accurate delay estimation is critical. Traditional correlation methods are limited by sampling frequency, achieving accuracy only at the sampling interval level. To achieve higher accuracy, subsample estimation is necessary. This letter proposes a precise delay calibration method using maximum likelihood iteration for subsample delay estimation. The proposed algorithm starts with the frequency domain features, first transforming the delay estimation problem into a phase estimation problem, and then calculating the likelihood function of the phase difference. A cost function is established based on the maximum likelihood criterion, and the optimal solution is obtained using the Newton iteration method. Compared to other algorithms, the proposed algorithm achieves superior accuracy in subsample delay estimation, meeting stringent calibration requirements in radio astronomy. Simulation and experimental results verify the validation of the algorithm.
{"title":"High-Precision Time Delay Calibration for Radio Astronomy Radars Based on Maximum Likelihood Iteration","authors":"Quanhua Liu;Bowen Cai;Xinliang Chen;Rui Zhu;Zhennan Liang","doi":"10.1109/LGRS.2025.3549788","DOIUrl":"https://doi.org/10.1109/LGRS.2025.3549788","url":null,"abstract":"In the calibration of distributed radar for radio astronomy, deep space radio sources are commonly used as calibration sources to correct interarray delay errors, and accurate delay estimation is critical. Traditional correlation methods are limited by sampling frequency, achieving accuracy only at the sampling interval level. To achieve higher accuracy, subsample estimation is necessary. This letter proposes a precise delay calibration method using maximum likelihood iteration for subsample delay estimation. The proposed algorithm starts with the frequency domain features, first transforming the delay estimation problem into a phase estimation problem, and then calculating the likelihood function of the phase difference. A cost function is established based on the maximum likelihood criterion, and the optimal solution is obtained using the Newton iteration method. Compared to other algorithms, the proposed algorithm achieves superior accuracy in subsample delay estimation, meeting stringent calibration requirements in radio astronomy. Simulation and experimental results verify the validation of the algorithm.","PeriodicalId":91017,"journal":{"name":"IEEE geoscience and remote sensing letters : a publication of the IEEE Geoscience and Remote Sensing Society","volume":"22 ","pages":"1-5"},"PeriodicalIF":0.0,"publicationDate":"2025-03-10","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143706708","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}