IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing最新文献

{"title":"Hybrid Integrated Feature Fusion of Handcrafted and Deep Features for Rice Blast Resistance Identification Using UAV Imagery","authors":"Peng Zhang;Zibin Zhou;Huasheng Huang;Yuanzhu Yang;Xiaochun Hu;Jiajun Zhuang;Yu Tang","doi":"10.1109/JSTARS.2025.3543190","DOIUrl":"https://doi.org/10.1109/JSTARS.2025.3543190","url":null,"abstract":"Nowadays, the combination of UAV remote sensing and deep learning has facilitated effective high-throughput field phenotyping for rice-blast-resistant breeding. However, breeding practices can hardly provide sufficient samples for each category due to the limitations of germplasm resources, which may cause data insufficiency and class imbalance. In addition, foliar and lesion details are often difficult to identify in UAV images due to the limitation of spatial resolution. As a result, the application of deep learning can lead to overfitting, as the model may struggle to acquire discriminative features. While previous studies have attempted to combine handcrafted and deep features to address problems with data insufficiency and class imbalances, image degradation still prevents the network from learning efficient representations for disease identification. To address these issues, this article proposes a hybrid integrated feature fusion (HIFF) method, in which a novel handcrafted-design-guided convolutional neural network module was employed to alleviate the problem of image degradation. Both handcrafted and deep learning branches were integrated in an end-to-end structure and applied to rice blast resistance identification. The proposed method was carefully evaluated using an ablation study, and the comparisons with state-of-the-art deep learning and feature fusion methods were conducted to demonstrate its superiority. Experimental results showed that the HIFF model outperformed mainstream methods by 0.0353 in F1-score and 0.0488 in accuracy on the practical rice-blast-resistant breeding applications. As such, the proposed method could be used to accelerate the process of rice-blast-resistant breeding.","PeriodicalId":13116,"journal":{"name":"IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing","volume":"18 ","pages":"7304-7317"},"PeriodicalIF":4.7,"publicationDate":"2025-02-18","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=10891690","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143655089","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"地球科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"MSTCNet: Toward Generalization Improving for Multiframe Infrared Small Target Detection","authors":"Ruining Cui;Na Li;Junfu Liu;Huijie Zhao","doi":"10.1109/JSTARS.2025.3542617","DOIUrl":"https://doi.org/10.1109/JSTARS.2025.3542617","url":null,"abstract":"Multiframe infrared small target detection plays an important role in various fields, especially in remote sensing. In continuous-frame infrared small target videos, factors such as the background change with the movement of the target. These changes lead to differences between the data distribution in actual application scenarios and the training scenarios. Existing deep learning methods are mostly designed for fixed scenarios. When facing scenarios with complex backgrounds and diverse changes, the generalization performance of the model is insufficient, leading to a decrease in detection accuracy and an increase in false alarms rate. To solve the problems mentioned above, combining the concept of domain generalization (DG) in transfer learning, we propose a multiscale spatio-temporal feature combined network (MSTCNet). First, we utilize the advantages of convolutional neural networks and recurrent neural networks, integrating them to build a high-performance structure. In addition, to further enhance generalization performance, we designed a selective physical information fusion (SPIF) module based on domain-invariant representation learning. This module enhances domain-invariant infrared small target features and reduces the impact of other irrelevant interferences. By integrating wavelet transform within the neural network, along with spatial attention and contrastive learning, SPIF strengthens domain-invariant features crucial for the task. Finally, in the experimental part, we adopt the DG verification method, dividing the dataset into different source domains and target domains for experimental verification. We verified the generalization performance of the proposed MSTCNet on two different datasets (IDGA and DTBA), and the experimental results confirmed the practicality and effectiveness of our method.","PeriodicalId":13116,"journal":{"name":"IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing","volume":"18 ","pages":"8416-8437"},"PeriodicalIF":4.7,"publicationDate":"2025-02-17","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=10891484","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143726486","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"地球科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"MCDiff: A Multilevel Conditional Diffusion Model for PolSAR Image Classification","authors":"Qingyi Zhang;Xiaoxiao Fang;Tao Liu;Ronghua Wu;Liguo Liu;Chu He","doi":"10.1109/JSTARS.2025.3542952","DOIUrl":"https://doi.org/10.1109/JSTARS.2025.3542952","url":null,"abstract":"With the swift advancement of deep learning, significant strides have been made in polarimetric synthetic aperture radar (PolSAR) image classification, particularly with the advent of diffusion models that allow for explicit class probability modeling. However, existing diffusion models have yet to fully leverage the rich polarimetric characteristics of PolSAR images. To address this, we propose the multilevel conditional diffusion (MCDiff) model for PolSAR image classification, incorporating three key strategies. First, a prior learning module is constructed to capture scattering characteristics across all three polarization basis parameter spaces, providing conditional guidance for the diffusion model. Second, a multiscale and multidimensional noise prediction module is designed to reduce the information loss when noisy labels and image features of different dimensions are fused to predict noise. Finally, a multilevel high-order statistical feature learning module is introduced to aid in the additive Gaussian noise prediction of noisy labels while mitigating the impact of PolSAR images' multiplicative speckle noise on the prediction. Experimental results on three benchmark datasets confirm MCDiff's ability to achieve high-performance explicit class probability modeling for PolSAR images among the compared methods.","PeriodicalId":13116,"journal":{"name":"IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing","volume":"18 ","pages":"6721-6737"},"PeriodicalIF":4.7,"publicationDate":"2025-02-17","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=10891635","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143594447","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"地球科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"A Multimodal Semantic Segmentation Framework for Heterogeneous Optical and Complex SAR Data","authors":"Sining Xiao;Peijin Wang;Wenhui Diao;Kun Fu;Xian Sun","doi":"10.1109/JSTARS.2025.3542487","DOIUrl":"https://doi.org/10.1109/JSTARS.2025.3542487","url":null,"abstract":"The advancement of remote sensing technology has led to a progressive enhancement in the resolution of remote sensing data, offering a multiperspective approach to Earth observation and facilitating a more comprehensive scene interpretation. As two most commonly utilized data sources in remote sensing, optical images, and synthetic aperture radar (SAR) data can provide complementary information, effectively compensating for the limitations inherent to a single modality. However, existing methods for using these two data sources face the following issues. First, insufficient utilization of the complete information provided by the source data. Second, inadequate consideration of the distinct characteristics of different modalities during feature extraction. Third, ignoring the misalignment between heterogeneous data, leading to large information loss. To tackle these challenges, we initially construct a benchmark dataset comprising complex-valued SAR data and optical images, named Multi-Complex-Seg. In order to fully mine the complete and valid information provided by both data sources, we construct a multimodal segmentation framework built on the theory of “subdomain extraction and cross-domain fusion,” in which we design a more suitable feature extractor for complex-valued SAR data, fully considering the unique geometric properties. In addition, a dynamic feature alignment module (DFAM) is proposed to further adjust the cross-modal features, and Cross-modal heterogeneous feature fusion module (CHFFM) first maps features into the same latent space to obtain better fused features. Both DFAM and CHFFM together reduce the huge semantic gap between modalities, thus facilitating the extraction of intramodal specificity and cross-modal complementarity. Extensive experiments on the proposed Multi-Complex-Seg confirm the effectiveness of our framework in comparison to other state-of-the-art multimodal segmentation approaches.","PeriodicalId":13116,"journal":{"name":"IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing","volume":"18 ","pages":"8083-8098"},"PeriodicalIF":4.7,"publicationDate":"2025-02-17","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=10891164","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143706816","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"地球科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Ocean Surface Retracking in Tropical Cyclones With the KaIA Airborne Radar Altimeter","authors":"Clayton M. Bjorland;Zorana Jelenak;Joseph W. Sapp;Casey G. Shoup;Bradley M. Isom;Paul S. Chang;James R. Carswell","doi":"10.1109/JSTARS.2025.3542635","DOIUrl":"https://doi.org/10.1109/JSTARS.2025.3542635","url":null,"abstract":"KaIA is an airborne Ka-band radar altimeter capable of centimetric range resolution and near real-time significant wave height retrievals. Beginning in 2019, KaIA has been installed on National Oceanic and Atmospheric Administration (NOAA) WP-3D Hurricane Hunter aircraft to collect data in Atlantic tropical cyclones during the hurricane season, and extratropical cyclones during the winter storm season. This article details recent retracker algorithm innovations that address specific difficulties with airborne altimetry in extreme weather. Our two most significant contributions to retracker algorithm development are: 1) a higher-order expansion of the classic Brown (1977) altimetry waveform to accommodate off-nadir pointing angles up to 3.25<inline-formula><tex-math>${}^{circ }$</tex-math></inline-formula>; 2) a GPS stabilization algorithm to enable along-track averaging while aircraft altitude is changing unpredictably. Comparisons against coincident measurements and modeled data are presented to validate algorithm improvements and document KaIA's performance throughout the 2021-2023 hurricane seasons. We measure less than 0.1 m bias in significant wave height retrievals relative to coincident satellite altimeters.","PeriodicalId":13116,"journal":{"name":"IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing","volume":"18 ","pages":"6480-6491"},"PeriodicalIF":4.7,"publicationDate":"2025-02-17","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=10891256","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143594353","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"地球科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"A Dual-Strategy Learning Framework for Hyperspectral Image Super-Resolution","authors":"Shuying Li;Ruichao Sun;San Zhang;Qiang Li","doi":"10.1109/JSTARS.2025.3542766","DOIUrl":"https://doi.org/10.1109/JSTARS.2025.3542766","url":null,"abstract":"Hyperspectral image super-resolution (HSI SR) has achieved remarkable success with deep neural networks. Currently, most methods in HSI SR assume a predetermined degradation model during training to synthesize low-resolution images. These methods falter when confronted with HSI exhibiting degradation patterns and their limited flexibility restricts practical application. In addition, these methods focus on the complex network designs for superior performance, which entail high resource consumption and limit their broad application. To address these issues, in this article, we propose a dual-strategy learning framework exploring meta-transfer learning for HSI blind SR. This framework can be applied to any SR network and facilitate performance enhancement. First, we pretrain a three-channel SR model on natural image data to address the issue of insufficient HSI data. Furthermore, we innovatively propose a transfer scheme, which directly applies our pretrained three-channel SR model to HSI, thereby significantly enhancing the spectral fidelity. To enhance the model's performance under specific degradation conditions, we incorporate meta-learning, enabling it to adapt to input images after a few iterations. Besides, we introduce attention-based knowledge distillation to equip our final network with the implicit representation capability of a meta network under a lightweight premise. Extensive experiments on three benchmark datasets demonstrate that the proposed approach outperforms existing methods in various degradations.","PeriodicalId":13116,"journal":{"name":"IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing","volume":"18 ","pages":"7480-7494"},"PeriodicalIF":4.7,"publicationDate":"2025-02-17","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=10891577","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143645331","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"地球科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"GANFlow: A Hybrid Model for SAR Image Target Open-Set Recognition Based on GAN and the Flow-Based Module","authors":"Jikai Qin;Jiusheng Han;Zheng Liu;Lei Ran;Rong Xie;Tat-Soon Yeo","doi":"10.1109/JSTARS.2025.3542738","DOIUrl":"https://doi.org/10.1109/JSTARS.2025.3542738","url":null,"abstract":"Most synthetic aperture radar (SAR) automatic target recognition (ATR) methods can achieve good recognition results only under the closed-set assumption. However, in practical applications, ATR models are often exposed to open environments, the general closed-set method may misclassify unknown categories as known categories, which is not reasonable. To tackle this issue, this article proposes an end-to-end hybrid model for SAR image open-set recognition (OSR), named GANFlow, which combines a generative adversarial network (GAN) with a flow-based module. The GANFlow achieves accurate classification of known categories and effective rejection of unknown categories. In this model, a classifiable convolution GAN is first designed to complete the training of the feature extraction module and classifier. Through adversarial training, the generated images enrich the training samples, which improves the ability of feature extraction and classification of the discriminator. Then, to find the difference in the probability density distribution of the extracted features, a flow-based module is adopted. Also, the features avoid interference from irrelevant background information in SAR images. Furthermore, by establishing an appropriate threshold, unknown categories can be efficiently rejected. Finally, the outputs of the classifier and the flow-based module are combined to complete the OSR of the SAR image target. The experimental results on the MSTAR and OpenSARShip public-measured datasets verify the robustness and generalization of the proposed method.","PeriodicalId":13116,"journal":{"name":"IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing","volume":"18 ","pages":"7083-7099"},"PeriodicalIF":4.7,"publicationDate":"2025-02-17","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=10891547","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143706814","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"地球科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"The Data-Optimized Oblique Mercator Projection","authors":"Sebastian von Specht;Malte J. Ziebarth","doi":"10.1109/JSTARS.2025.3542802","DOIUrl":"https://doi.org/10.1109/JSTARS.2025.3542802","url":null,"abstract":"Map projections transform the Earth's curved surface into a plane and are thus crucial for mapping and geospatial analysis. However, projections inevitably introduce distortion, requiring the selection of a suitable map projection for the mapped region. The conventional approach is to choose from predefined map projections. Unfortunately, the available projections are limited in variety and can be difficult to evaluate effectively. We propose an alternative approach: rather than selecting from a predefined set of projections, we introduce an algorithm that optimizes a single projection for a given dataset: Data-Optimized Oblique Mercator (DOOM). At its core is the HOM projection, featuring a flexible set of adjustable parameters and a universal implementation in GIS platforms and related software. DOOM utilizes the well-established optimization algorithms Levenberg–Marquardt, Adamax, and BFGS, to optimize the projection parameters, minimizing distortion in the mapping of geospatial data. The algorithm supports various objective functions (e.g., <inline-formula><tex-math>$L^{1}$</tex-math></inline-formula>- and <inline-formula><tex-math>$L^{2}$</tex-math></inline-formula>-norms, minmax) and can be extended to incorporate data weighting. The methodology is validated through several case studies, highlighting its adaptability across diverse applications. In addition, we introduce a GIS plugin to streamline the use of optimized projection parameters, enhancing accessibility for the geospatial community.","PeriodicalId":13116,"journal":{"name":"IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing","volume":"18 ","pages":"6916-6939"},"PeriodicalIF":4.7,"publicationDate":"2025-02-17","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=10891375","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143706706","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"地球科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"A New Fast Sparse Unmixing Algorithm Based on Adaptive Spectral Library Pruning and Nesterov Optimization","authors":"Kewen Qu;Fangzhou Luo;Huiyang Wang;Wenxing Bao","doi":"10.1109/JSTARS.2025.3541257","DOIUrl":"https://doi.org/10.1109/JSTARS.2025.3541257","url":null,"abstract":"In recent years, hyperspectral sparse unmixing (HSU) has garnered extensive research and attention due to its unique characteristic of not requiring the estimation of endmembers and their number. However, the high coherence and large-scale nature of the prior spectral library frequently lead to substantial computational costs and limited unmixing accuracy in the optimization model, thereby hindering the efficiency and further promotion of HSU in practical engineering applications. To address these shortcomings, this article proposes a new fast two-step sparse unmixing algorithm, called NeSU-LP, which is based on adaptive spectral library pruning technology and the Nesterov fast optimization strategy. In this method, HSU is divided into two independent and consecutive subprocesses: coarse unmixing and fine unmixing. Specially, first, in the coarse unmixing stage, we design a sparse optimization model based on the initial large spectral library, requiring only a few iterations to initially estimate the row-sparse abundance matrix. Subsequently, the proposed atomic (i.e., endmember) activity evaluation method is utilized to screen the active endmembers, analyze the abundance matrix, and prune the endmembers in the spectral library. Irrelevant endmembers are removed, reducing the spectral library size and generating a low-coherence, small-scale endmember matrix. Finally, in the fine unmixing process, we retain the effective atomic abundance rows obtained in the previous stage and design the final fine hyperspectral unmixing model based on the pruned, small-scale endmember matrix. In addition, to enhance the smoothness of the abundance maps, graph Laplacian regularization is introduced during the fine unmixing stage. The Nesterov fast gradient strategy is employed to accelerate the iterative process of fine unmixing, ultimately achieving second-order convergence efficiency for the algorithm. Numerous experiments were conducted on both synthetic and real datasets, comparing them with state-of-the-art methods. The experimental results demonstrate the high efficiency and advancement of the proposed method.","PeriodicalId":13116,"journal":{"name":"IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing","volume":"18 ","pages":"6134-6151"},"PeriodicalIF":4.7,"publicationDate":"2025-02-14","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=10887113","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143564201","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"地球科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Joint Correlations Sparse Bayesian Learning STAP With Prior Knowledge of Clutter Ridge","authors":"Junhao Cui;Zhangxin Chen;Jing Liang","doi":"10.1109/JSTARS.2025.3542421","DOIUrl":"https://doi.org/10.1109/JSTARS.2025.3542421","url":null,"abstract":"Space-time adaptive processing (STAP) based on sparse Bayesian learning (SBL) can significantly improve clutter suppression performance utilizing clutter sparsity. However, the existing SBL-STAP algorithms lack full use of correlations, which leads to unsatisfactory performance and slow convergence speed. In this article, we propose a joint correlations SBL-STAP (JCSBL-STAP) algorithm to improve clutter suppression performance. It comes from a rational idea that the clutter ridge in the space-time domain is not only the origin of clutter sparsity, but also the origin of correlations. Normally, the amplitude of scatterers along the clutter ridge are correlated between multiple samples and have clustered correlation properties in each sample. The JCSBL-STAP algorithm utilizes a joint correlations sparse prior to exploiting both correlations and provides a multisample correlation decoupling framework to update hyperparameters. The algorithm is executed on a proposed hybrid prior dictionary. Compared with the conventional uniform dictionary, the hybrid prior dictionary can easily express the clustered correlation properties and effectively alleviates the off-grid problem. Experimental results confirm the performance of the proposed method on both simulated data and measured Mountain-Top data.","PeriodicalId":13116,"journal":{"name":"IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing","volume":"18 ","pages":"6820-6832"},"PeriodicalIF":4.7,"publicationDate":"2025-02-14","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=10890973","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143601888","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"地球科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}