Qiao Cheng;Xiangbo Gong;Bin Hu;Hongyu Zhu;Zhiyu Cao
{"title":"Seismic Data Sparse Representation Using Swin Transformers","authors":"Qiao Cheng;Xiangbo Gong;Bin Hu;Hongyu Zhu;Zhiyu Cao","doi":"10.1109/LGRS.2024.3510685","DOIUrl":"https://doi.org/10.1109/LGRS.2024.3510685","url":null,"abstract":"Seismic data preprocessing significantly benefits from advanced sparse representation and domain transformation techniques to enhance denoising, wavefield separation, and data reconstruction. This study introduces a novel approach utilizing a deep learning framework for discrete sparse representation of seismic data. Our method utilizes a Swin Transformer-based encoding-decoding framework, which combines the hierarchical structures of CNNs with the self-attention mechanism of Transformers, to model both local and global information efficiently. This integration enables the precise characterization of seismic reflection events and the reconstruction of seismic records from a constructed sparse feature space. The proposed model has been rigorously tested on both simulated and field datasets, demonstrating its robustness, and potential provides superior decomposition of seismic data.","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":"2024-12-04","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142810452","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}
{"title":"Three-Dimensional Joint Inversion of Borehole and Surface Magnetic Data of the Cangyi Iron Mine, Shandong Province (East China)","authors":"Jiantai Zhang;Hecai Cao;Chenghe Zhu;Xiange Jian;Yongsheng Sun;Yu Li;Changsheng Guo;Liwei Yuan;Lei Yu;Xianfu Du","doi":"10.1109/LGRS.2024.3509415","DOIUrl":"https://doi.org/10.1109/LGRS.2024.3509415","url":null,"abstract":"The Cangyi iron ore belt is a geologically significant sedimentary-metamorphic iron ore belt in China. The transition from open-pit mining to underground mining in the belt is primarily driven by the depletion of surface resources, necessitating the exploration of deep-seated deposits. However, the distribution of these deposits in the Cangyi iron ore belt is intricately controlled by basement fold structures and exhibits late-stage modifications, posing challenges for accurate evaluations of the ore deposits. This study uses a surface magnetic survey to map the planar distribution of the ore bodies. Via 3-D joint inversion of borehole and surface magnetic data, the planar and deep-seated distribution of magnetic iron ore in the belt is obtained. The joint inversion of surface and borehole magnetic data enhances the vertical and horizontal resolutions of deep-seated magnetic sources. This approach is a crucial geophysical method for exploring and characterizing deep mineral resources. Reliance on surface structural traces alone is insufficient for accurate reconstructions of deep structures. Using magnetite inferred from the joint inversion as a marker layer and analyzing the structural trace of the deposit is essential to evaluating the deposit. Previous speculation regarding the presence of a deep-seated ore body at a depth of 1400 m in the exploration area is dispelled by joint inversion of borehole-surface magnetic data, which reveals the absence of highly magnetized magnetic bodies at a depth of 1300 m. This finding provides a clear direction for further deep exploration.","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":"2024-12-04","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142825932","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}
{"title":"FedShip: Federated Learning for Ship Detection From Multi-Source Satellite Images","authors":"Anh-Kiet Duong;Tran-Vu La;Hoàng-Ân Lê;Minh-Tan Pham","doi":"10.1109/LGRS.2024.3511122","DOIUrl":"https://doi.org/10.1109/LGRS.2024.3511122","url":null,"abstract":"Detecting ships from satellite imagery is vital for maritime surveillance. Most current methods rely on deep learning (DL), which requires a large number of high-quality annotated images to train accurate models. Since satellite imagery comes from various sensors, DL-based ship detection algorithms need to perform well across different sensor types. However, privacy concerns, especially with commercial images, limit data, and annotation sharing. Federated learning (FL) offers a promising solution for collaborative learning while addressing these concerns. Despite its potential, research on FL for ship detection is still sparse. This study implements and evaluates three FL models for detecting ships using multi-source optical satellite images, spanning high to low resolution. Our experiments on two distinct datasets demonstrate that FL models significantly enhance detection performance without centralizing data. Source codes are publicly available at \u0000<uri>https://github.com/ffyyytt/FLYOLO</uri>\u0000.","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":"2024-12-04","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142858945","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}
{"title":"FDTD Medium Dimension Selection Guidelines for GPR Synthetic Data Generation","authors":"Noushin Khosravi Largani;Seyed Zekavat;Himan Namdari","doi":"10.1109/LGRS.2024.3510683","DOIUrl":"https://doi.org/10.1109/LGRS.2024.3510683","url":null,"abstract":"Ground-penetrating radar (GPR) has been traditionally used for subsurface assessment. In many applications, such as precision agriculture via drone-borne radar, it is critical to use machine learning (ML) techniques to map GPR received signals into soil subsurface moisture and texture. Supervised ML methods need a large number of labeled data for their training process which is expensive and time-consuming to attain through actual field measurements. The gprMax software, which is created based on the finite difference time domain (FDTD) method, has been introduced as a reliable tool to emulate soil media and create synthetic labeled data. Proper selection of gprMax soil medium dimensions is critical to the generation of reliable synthetic data. The selection of large soil medium dimensions for gprMax emulations leads to synthetic data consistent with realistic scenarios. However, larger medium dimensions lead to higher computation complexity. This letter investigates and validates a proper selection of medium dimensions that maintains a tradeoff across the accuracy and computational complexity of creating synthetic data. The results of this study are critical to researchers who adopt gprMax or any FDTD-oriented emulations for soil subsurface assessment. To maintain a tradeoff between accuracy and complexity, the letter confirms that the minimum medium surface dimension should be in the order of 1.5 times the maximum wavelength.","PeriodicalId":91017,"journal":{"name":"IEEE geoscience and remote sensing letters : a publication of the IEEE Geoscience and Remote Sensing Society","volume":"22 ","pages":"1-4"},"PeriodicalIF":0.0,"publicationDate":"2024-12-04","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142844389","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}
Yanqiong Liu;Sen Lei;Nanqing Liu;Jie Pan;Heng-Chao Li
{"title":"Memory-Augmented Differential Network for Infrared Small Target Detection","authors":"Yanqiong Liu;Sen Lei;Nanqing Liu;Jie Pan;Heng-Chao Li","doi":"10.1109/LGRS.2024.3510803","DOIUrl":"https://doi.org/10.1109/LGRS.2024.3510803","url":null,"abstract":"Traditional U-Net-based methods in infrared small target detection (IRSTD) have demonstrated good performance. However, they often struggle with challenges such as blurred contour and strong interference in complex backgrounds. To overcome these issues, we propose a memory-augmented differential network (MAD-Net), which integrates two key modules: the adaptive differential convolution module (AdaDCM) and the memory-augmented attention module (MemA2M). AdaDCM leverages multiple differential convolutions to capture detailed edge information, with an adaptive fusion mechanism to weight and aggregate these features. In the deeper layers, by introducing the dataset-level representations through a learnable memory bank (LMB), MemA2M can enhance current features and effectively mitigate background interference. Extensive experiments on four public IRSTD datasets demonstrate that MAD-Net outperforms state-of-the-art methods, showcasing its superior capability in handling complex scenarios. The code is available at: \u0000<uri>https://github.com/joan2joan/MAD-Net</uri>\u0000.","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":"2024-12-04","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142810202","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}
{"title":"Enhanced Electrical Resistivity Tomography With Prior Physical Information","authors":"Zhuo Jia;Meijia Huang;Zhijun Huo;Yabin Li","doi":"10.1109/LGRS.2024.3505607","DOIUrl":"https://doi.org/10.1109/LGRS.2024.3505607","url":null,"abstract":"Electrical resistivity tomography (ERT) is a key geophysical technique that provides detailed information on subsurface structures by measuring the distribution of electrical resistivity underground. ERT suffers from limitations in electrode arrangement, interference from environmental and instrument noise, and existing data processing algorithms that fail to adequately consider geological heterogeneity and uncertainty, resulting in insufficient inversion resolution. Traditional ERT methods rely on simplified algorithms and a limited number of observation points, which smooths model details and further reduces resolution. To address the resolution issues in ERT, this article proposes a deep learning inversion method that integrates prior physical information. This method uses low-resolution inversion results as prior knowledge to provide the deep learning algorithm with a constrained initial model, thereby combining the physical basis of traditional methods with the data-driven advantages of deep learning. The method not only retains the strengths of traditional inversion but also enhances the resolution and imaging efficiency of the inversion model using deep learning technology. Synthetic data experiments demonstrate that integrating deep learning significantly improves the model’s ability to detail subsurface structures, especially in the transition zones of shallow structures and the recovery of deep anomalies. Results from measured data indicate that the proposed method not only achieves high-resolution inversion but also maintains good consistency with prior information.","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":"2024-12-04","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142797980","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}
Zhaolong Wang;Xiaokuan Zhang;Weike Feng;Xixi Chen;Ninghui Li
{"title":"Deep Unfolded Atomic Norm Minimization Algorithm for Space-Time Adaptive Processing","authors":"Zhaolong Wang;Xiaokuan Zhang;Weike Feng;Xixi Chen;Ninghui Li","doi":"10.1109/LGRS.2024.3509520","DOIUrl":"https://doi.org/10.1109/LGRS.2024.3509520","url":null,"abstract":"As an effective clutter suppression method for airborne radar, the atomic norm minimization (ANM)-based space-time adaptive processing (STAP) method suffers from high computational complexity and parameter setting difficulty. To solve these problems, a deep unfolded (DU) ANM algorithm is proposed for STAP in this study. First, the clutter estimation problem based on ANM is established. Then, the problem is solved via the alternating direction method of multipliers (ADMMs) and a deep neural network (DNN), which is trained by designing an appropriate loss function and constructing a complete dataset. At last, the clutter-plus-noise covariance matrix (CNCM) and the STAP weighting vector are obtained by processing the training range cell data via the trained network. Simulation results show that the proposed DU-ANM-STAP method can achieve higher clutter and noise suppression performance with lower computational cost than the existing ANM-STAP methods.","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":"2024-12-02","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142821269","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}
{"title":"Implicit Neural Attention for Removing Blur in Remote Sensing Images","authors":"Yaowei Li;Hanmei Yang;Xiaoxuan Chen;Hang An;Bo Jiang","doi":"10.1109/LGRS.2024.3509894","DOIUrl":"https://doi.org/10.1109/LGRS.2024.3509894","url":null,"abstract":"Deblurring in remote sensing images is a challenging task due to the long-range imaging capabilities of remote sensing sensors, which often results in image blur. Factors contributing to image blur include atmospheric disturbances during long-range imaging or the orbital motion of remote sensing platforms. The existing methods remove blur in remote sensing images using the traditional attention mechanism, which focuses on a limited number of features. However, they often overlook the features among neighboring positions in blurry areas, and these areas contain more relevant features. Leveraging these features can effectively assist in restoring the complex object textures of remote sensing blurry images. To achieve this, we propose a novel implicit neural attention mechanism for assembling more relevant features implied by surrounding coordinates. Specifically, we use the features and their corresponding coordinates to learn the enhanced feature representation with more relevant features, and this representation can be used to derive the deblurred images. Extensive experiments demonstrate that our proposed method, INA-RSDeblur, outperforms the state-of-the-art deblurring methods in remote sensing blurry images.","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":"2024-12-02","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142810204","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}
{"title":"Cross-Domain Density Map-Generated Ship Counting Network for Remote Sensing Image","authors":"Yaxiong Chen;Qijian Li;Kai Yan;Shengwu Xiong","doi":"10.1109/LGRS.2024.3510093","DOIUrl":"https://doi.org/10.1109/LGRS.2024.3510093","url":null,"abstract":"In recent years, with the continuous development of remote sensing technology, maritime ship monitoring has become an important research area. Accurately counting the number of ships in remote sensing images is crucial for maritime traffic safety, fisheries management, and marine environmental protection. Existing methods typically use Gaussian kernel functions to generate density maps; however, due to the varied shapes of ships that do not conform to the Gaussian kernel, the resulting density maps fail to accurately reflect the true forms of ships, thereby affecting counting performance. To overcome these limitations, we introduce the cross-domain density map-generated ship counting network (CDDMNet). This network innovatively incorporates a cross-domain feature fusion module (CDFFM), which effectively adapts to ships of varying sizes and shapes. In addition, we have introduced the feature correlation regularization constraint (FCRC) and the integrated loss function, which effectively overcome the disturbances that may arise from variations in ship sizes and enhance the model’s adaptability to changes in ship types and environmental conditions. Experimental results show that the CDDMNet has achieved excellent performance across multiple remote sensing image datasets. Finally, on the RSOC dataset, the mean absolute error (MAE) reached 52.80 and the root mean squared error (RMSE) reached 69.77.","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":"2024-12-02","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142810205","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}
Hai Huang;Shengqi Zhu;Xiongpeng He;Ximin Li;Guisheng Liao
{"title":"An Improved Time Diversity HRWS Imaging Method Based on Transmit Waveform Optimization Design","authors":"Hai Huang;Shengqi Zhu;Xiongpeng He;Ximin Li;Guisheng Liao","doi":"10.1109/LGRS.2024.3510380","DOIUrl":"https://doi.org/10.1109/LGRS.2024.3510380","url":null,"abstract":"This letter proposes a time-diverse wide-swath imaging radar transmit waveform optimization design method. First, based on the imaging geometry and zebra maps, we obtained the angles corresponding to the range occlusion zone. Then, using the mapping characteristics between the range frequency and beam scanning angle in time-diverse array (TDA) radar, as well as the occlusion region information, we performed a 2-D optimization design of the transmit waveform in the fast time and range frequency domain. Finally, the limited energy can be effectively skipped over the occlusion regions and flexibly allocated to the observable areas. Compared with the traditional TDA system, this method achieves a larger imaging swath and energy utilization efficiency. The effectiveness of the proposed method is verified through simulation experiments.","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":"2024-12-02","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142844516","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}