IEEE Transactions on Geoscience and Remote Sensing

This paper proposes a novel deep learning-based waveform inversion method that enables rapid and high-precision extraction of nonlinear waveform information from seismic data. To address critical challenges in existing deep learning full waveform inversion (FWI), such as poor network generalization and huge computational cost, we introduce angle-domain generalized Radon transform (AD-GRT). As int…

Historical maps are unique and valuable archives that document geographic features across different time periods. However, automated analysis of historical map images remains a significant challenge due to their wide stylistic variability and the scarcity of annotated training data. Constructing linked spatio-temporal datasets from historical map time series is even more time-consuming and labor-…

Bridges are crucial components of urban infrastructure, requiring continuous monitoring to ensure safety and longevity. With the advancement of Integrated Sensing and Communication (ISAC), base stations (BSs) can be leveraged for cost-effective and real-time bridge health monitoring. However, the sway of the antenna feeder system (AFS) of the BS under wind disturbances introduces phase errors in …

Blended noise is coherent noise introduced during the simultaneous acquisition of multiple seismic sources. Due to its similar time-distance curve characteristics with effective waves in seismic records, suppressing blended noise poses great challenges. To address this problem, we propose a curvelet-transform-based method incorporating multi-level texture perception and regularization. Seismic da…

Hyperspectral image (HSI) classification remains challenging due to complex spectral–spatial patterns, the growing demand for energy-efficient modeling in resource-constrained remote sensing applications, and the structural limitations of existing graph convolutional network (GCN)- and spiking neural network (SNN)-based methods. Conventional GCNs rely on static graphs and coarse neighborhood aggr…

Remote sensing video referring object segmentation (RS-RVOS) is challenged by weak target saliency and severe visual information truncation in dynamic scenes, making it extremely difficult to maintain discriminative target representations during segmentation. Moreover, progress in this field is hindered by the absence of large-scale dedicated benchmarks, while existing models are often affected b…

Interactive object detection paves a promising way to alleviate the annotation burden, particularly for size-limited instances in remote sensing images. While early studies emphasize fortifying object representations through click-based interactions, the alternative strategy of harnessing image as a prior to derive informative clicks remains largely under-explored—despite its substantial potentia…

Road extraction from remote sensing imagery has been widely applied in fields such as navigation systems, disaster response, and autonomous driving. However, existing deep learning-based models suffer from insufficient capability to extract curved and slender roads, as well as challenges in balancing model complexity and performance. This paper proposes a novel road extraction method based on 32-…

Geothermal anomalies, as surface manifestations of geothermal activity, are essential indicators of subsurface processes, and mapping them is a key step toward understanding the global distribution of geothermal potential. However, large-scale patterns remain poorly characterized due to limited ground-based observations and complicated environmental interferences. This study presents the first gl…

To address the challenges of imbalanced multi-scale feature rhythm, network representation defects, and insufficient loss constraints in the generation of infrared images converted from aerial visible light images, a diffusion bridge generation model focusing on multi-scale enhancement is proposed. Starting from the full-link design of multi-scale generation, the model, in terms of diffusion proc…

Cross-domain few-shot hyperspectral image classification (CDFSL HSIC) presents significant challenges, primarily due to spectral variability across domains and the scarcity of labeled target samples. To tackle these issues, we propose a Graph Relation Aggregated Spectral Perception Network (GRASP-Net), which progressively models spectral and structural information from the sample level to the cat…

Pan-sharpening aims to fuse the texture-rich details of a high-resolution panchromatic (PAN) image with the spectral information of a low-resolution multi-spectral (MS) image to generate a high-resolution multi-spectral counterpart. Recently, state-space models (SSMs), notably the Mamba architecture, have emerged as promising alternatives for global sequence modeling, achieving efficient processi…

To address the problem that current hyperspectral image super-resolution methods struggle to effectively extract spatial-spectral information, resulting in degraded performance, this paper proposes a dual-branch network for hyperspectral images super-resolution (DBNSR). In the proposed framework, one branch transforms hyperspectral image super-resolution into abundance map super-resolution throug…

Lane detection is a fundamental task in computer vision with broad applications in autonomous driving. Recent advances have focused on line-anchor-based representations for their efficiency and high accuracy. However, most existing methods adopt anchors that are continuous in point sampling yet discrete in width and fixed in direction. This leads to geometric misalignments with real-world lane li…

Existing deep learning-based cross-view geolocalization methods primarily focus on improving the accuracy of cross-domain image matching. Less attention is paid to ensuring that models can comprehensively capture contextual information around the target and minimize the cost of localization errors. To support quantitative research into this Distance-Aware Cross-View Geo-Localization (DACVGL) prob…

Rapid assessment of earthquake-damaged building-such as area, quantity, and severity-is crucial for timely post-earthquake rescue. Pre- and post-earthquake data collected from optical remote sensing systems can effectively resolve earthquake deformation. Among them, the normalized correlation coefficient reduction (NCCR) shows promise in directly reflecting building damage. Here we propose an aut…

Deep learning algorithms have driven substantial progress in remote sensing semantic segmentation. However, conventional approaches typically rely on predefined semantic categories, necessitating costly data annotation and model retraining when new classes are introduced. While large-scale vision-language models, such as CLIP, enable segmentation of arbitrary class with natural language guidance,…

Recent advances in optical remote-sensing have expanded the use of unmanned aerial vehicles (UAV), yet accurately localising dense, small objects in low-resolution UAV imagery remains challenging. We propose RSRT-DETR, a lightweight DETR derivative tailored for UAV-borne remote-sensing that jointly enhances global context modelling and fine-grained detail preservation. RSRT-DETR couples a Polar D…