Co-registration of multimodal remote sensing images (MRSI) remains a challenging task due to nonlinear radiometric differences (NRD) and geometric distortions. This letter proposes a robust registration method based on steerable Hermite filters and geometric constraints. The method consists of three stages: coarse registration, image matching, and fine registration. In the first stage, geo-referencing information is used to partition the overlapping region into sub-blocks for initial alignment. In the second stage, steerable Hermite filters are applied to extract multi-scale and multi-orientation structural features, which are fused into a principal moment map to detect highly repeatable feature points. A structural descriptor is subsequently constructed from an orientation index map generated through hierarchical pooling, enabling robust matching under NRD. In the final stage, an iterative optimization strategy incorporating geometric constraints is employed to eliminate mismatches and refine the registration. Experimental results on multiple MRSI datasets demonstrate that the proposed method outperforms state-of-the-art approaches in terms of both matching robustness and registration accuracy.

Seafloor seismic acquisition systems, such as ocean-bottom node (OBN) and cable (OBC) arrays, provide multi-component data essential for high-resolution imaging. Among them, vertical geophone recordings are often contaminated by shear-type noise (known as Vz noise), which arises from near-surface scattering and poor sensor coupling. This noise appears as incoherent energy in common shot gathers (CSGs) and as coherent hyperbolic events in common receiver gathers (CRGs), reducing data quality. We present a hybrid Radon transform-based method for attenuating shear noise in seafloor vertical geophone data. By combining linear and hyperbolic Radon transforms, the method enables effective separation of signal and noise. Two approaches are explored: (1) a geophone-only denoising in the CSG domain that reduces random noise and indirectly suppresses coherent noise in CRGs, and (2) a joint sparsity recovery method that incorporates hydrophone data as a reference to remove coherent noise in CRGs. Results on the Equinor Volve OBC dataset demonstrate effective noise attenuation and signal preservation, offering a practical solution for denoising multi-component seafloor seismic data.

We apply the self-supervised learning technique of vision transformers masked autoencoder (ViT MAE) models with the goal of to producing a feature extractor vision transformer (ViT) backbone for neural networks that receive seismic data as input. We then evaluate the quality of these backbones by coupling them to a simple linear prediction head and fine-tuning these models in a seismic semantic segmentation task. We compare domain-specific ViT MAE against cross-domain pretrained and randomly initialized ViTs, and show that it yields superior performance in low-data regimes. Furthermore, we also demonstrate that pretraining loss correlates with downstream performance, supporting its use as a proxy for feature quality.

Change detection (CD) networks commonly rely on spatial-domain operations, where convolutional or attentional filters implicitly mix spectral components and struggle to effectively distinguish structural changes from appearance variations. To address this limitation, we propose a Frequency-Domain Enhancement and Fusion Network (FDEF-Net). The core of FDEF-Net is a novel module that explicitly decomposes bi-temporal feature differences into amplitude and phase components, upon which a decoupled frequency-domain attention mechanism is applied. Furthermore, an adaptive dual-stream fusion strategy with learnable weights is designed to effectively integrate the frequency pathway with spatial features, enabling systematic and flexible cross-domain modeling. Experiments on three public benchmark datasets LEVIR-CD, WHU-CD, and GZ-CD demonstrate that this architecture consistently outperforms strong baselines under lightweight complexity. The source code of FDEF-Net will be released at https://github.com/Camille7911/FDEF.

Remote sensing semantic segmentation is a critical technology in the field of remote sensing image processing, with broad applications in environmental monitoring, urban planning, disaster assessment, and resource exploration. Despite the transformative impact of convolutional neural networks (CNNs) on this domain, CNN-based methods often encounter limitations due to their localized receptive fields, which struggle to capture the global context necessary for accurate segmentation in complex remote sensing imagery. In this letter, a novel approach is presented for remote sensing semantic segmentation using a mamba-based model named PPmamba. The PPmamba model integrates Resblock and PPmamba within an encoder-decoder framework to effectively capture both local and global contextual information from high-resolution remote sensing images. Leveraging the strengths of the Mamba architecture, our model employs selective scanning to efficiently process long sequences, overcoming the limitations of traditional CNNs and transformers in handling large-scale images with complex scenes. Extensive experiments on two benchmark datasets (Potsdam and Vaihingen) demonstrate the superiority of our PPmamba model against state-of-the-art models, achieving significant improvements in segmentation results. The codes will be available at <uri xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">https://github.com/Jerrymo59/PPMambaSeg</uri>.

Geosynchronous orbit (GEO) satellites are widely used in communications and navigation fields due to their special properties, such as comprehensive ground coverage and fixed subsatellite trajectories. The safety of GEO satellites can be ensured by using space-borne inverse synthetic aperture radar (SBISAR) to achieve high-precision observations or monitoring of uncatalogued targets near the GEO satellites. However, the amount of echo data is significant due to the slow relative motion between the GEO targets and SBISAR during long coherent processing intervals (CPIs). In this work, a variable sampling start echo receiving method is proposed to reduce the amount of redundant echo data, and the timing which combining the radar tracking with imaging is designed. First, the Earth center inertial (ECI) coordinate position vectors of the GEO target can be obtained through the monopulse method and coordinate transformation. Second, the initial orbit parameters of the GEO target are established by the Gibbs method. Third, the slant range of the GEO target is obtained by the interacting multiple model (IMM). Finally, the slant range is used to design the sampling start of the echo window and the timing of tracking imaging integration. Taking GEO satellite imaging as an example, experiments based on simulation data verify the effectiveness of the proposed method.

Synthetic Aperture Radar (SAR) represents nowadays a key technology in Earth Observation (EO), evolving its original capabilities into both large-scale monitoring of geophysical parameters and very high-resolution imaging with short revisit times. The increase in system performance and the wide range of application scenarios require significant efforts in the design of current and future SAR missions: one of the most critical on-board operations is the digitization of the received echoes, directly impacting the final image quality and, at the same time, limited by the available downlink capacity. State-of-the-art quantization methods, such as Block Adaptive Quantization (BAQ), offer a good trade-off between signal quality and overall complexity but lack adaptivity to the imaged scenario. This leads to different impacts of the quantization error on the final SAR image. As an evolution of BAQ, Performance-Optimized BAQ (PO-BAQ) is a recently proposed quantization method, which addresses this issue by employing variable quantization rates across the scene, targeting specific performance requirements in the final SAR image. In this paper, we present a feasibility study of variable bitrate allocation in a realistic SAR mission scenario: to ensure flexibility, we consider the bitrate allocation map (BRM) to be uploaded at commanding phase during each ground segment contact, individually tailoring the required performance quality for each acquisition. State-of-the-art uplink data rates are considered, and the complete performance evaluation after SAR processing is carried out using the experimental on-board processor developed within the SOPHOS Horizon 2020 project.

In recent years, contrastive learning (CL) methods have achieved remarkable success in few-shot classification of polarimetric synthetic aperture radar (PolSAR) images. However, existing CL models based on Euclidean metric typically vectorize PolSAR data into real-valued or complex-valued vectors with independent channels, disrupting the inherent correlations between polarimetric channels. To address this issue, we propose a novel CL framework, the Riemannian-Euclidean CL network (RECLNet). The proposed RECLNet allows direct input of polarimetric covariance matrices, overcoming the limitations of conventional Euclidean-based CL models that require vectorizing PolSAR data. First, RECLNet constructs hard positive samples by leveraging polarimetric information within the Riemannian space. Second, a series of carefully designed Riemannian manifold operation (RMO) layers are used to extract the Riemannian geometric features of PolSAR data while preserving its matrix structure. Finally, the vision transformer (ViT) is adopted as the backbone of the Euclidean metric encoding to capture contextual information. Experiments conducted on two widely used PolSAR datasets demonstrate that the proposed approach achieves superior performance compared with existing state-of-the-art methods.

This study proposes and benchmarks a lightweight pipeline for large-scale parcel-level crop classification using time-series data. The key contributions include a scalable SITS download strategy, a novel parcel-level crop classification dataset for the USA States of Texas and California, and a benchmark of classification techniques tested under varied data availability scenarios, highlighting the pipeline’s potential for enhancing agricultural monitoring systems. The methodology extracts basic descriptive statistics integrating with agricultural and cloud filtering, leveraging field boundaries delineation and USDA Cropland Data Layer datasets. Experimental results show that pretrained and fine-tuned models, such as SITS-BERT, outperform Random Forest and Support Vector Machine approaches, achieving an F1 score of 98.2% and overall accuracy of 99.3% on the Texas dataset for the more abundant training data scenarios. The pipeline has a projected computational speed-up of at leastat least 2, 500× compared to pixel-based methods for the tested datasets. The download pipeline is available on AgriGEE.lite library in https://pypi.org/project/agrigee-lite. The full benchmark results and related code are available at https://github.com/mateuspinto/light-crop-classification.

Multibaseline and tomographic synthetic aperture radar (SAR) data are often affected by phase distortions known as phase screens. These distortions stem either from atmospheric effects or residual errors in platform motion. Calibrating and compensating for the phase screen is crucial to prevent spreading and defocusing in multidimensional tomographic imaging. Given the growing interest in artificial intelligence and deep learning, we aim to utilize their potential to develop a phase calibration process for SAR tomographic data. Our proposed framework is based upon a convolutional neural network (CNN) and generates training patches directly from the tomographic images under consideration, without relying on external references or resources. Once trained, the network effectively estimates phase distortions across the entire image; these are then used to calibrate the tomographic data. Experimental results from AfriSAR and UAVSAR tomographic datasets are included to showcase the effectiveness of the proposed solution.