Aug 2026· Italian National Conference on Sensors· Vol 26· 0 citations· 40 references
Medicine
TL;DR
A novel SR framework based on the flow matching paradigm and a diffusion transformer, named FlowT-SR, which achieves superior and reliable reconstruction quality by jointly mitigating sensor noise and thin cloud interference, achieving superior reconstruction performance compared with current state-of-the-art methods in terms of both PSNR and SSIM.
Abstract
Remote sensing image super-resolution (SR) aims to enhance spatial resolution and recover image details, which typically enhances the quality of optical remote sensing imagery. However, interference from cloud haze cover and sensor noise often leads to distorted details and artifacts in reconstructed images of conventional deep learning SR approaches, significantly limiting reconstruction fidelity. To address these challenges, we propose a novel SR framework based on the flow matching paradigm and a diffusion transformer, named FlowT-SR, which achieves superior and reliable reconstruction quality by jointly mitigating sensor noise and thin cloud interference. First, an evolution path from low-resolution images to ground-truth images is constructed based on the optimal transport displacement interpolation mechanism, and the corresponding vector field that governs this evolution is employed as the supervision signal for subsequent model training. Then, a multi-scale interference suppression (MSIS) module is combined with a novel diffusion transformer network (DiTNet) to predict the vector field. The MSIS module performs preliminary denoising and captures the spatial distribution of thin cloud and haze in low-resolution images, providing degradation-aware feature representations for DiTNet. Subsequently, a DiTNet is presented to predict the evolution vector field obtained in the first stage, which consists of ten layers based on the diffusion transformer. By accurately predicting the vector field at any time step, the model effectively reduces the impact of cloud haze and noise interference to improve the reconstruction precision. Finally, driven by the predicted vector field along the evolution path, the SR remote sensing image is generated through solving the corresponding ordinary differential equation, yielding cloud-free and noise-reduced results. Extensive experiments on our dataset and the public CUHK Cloud Removal dataset demonstrate that FlowT-SR effectively suppresses cloud haze and noise interference, achieving superior reconstruction performance compared with current state-of-the-art methods in terms of both PSNR and SSIM.
This work proposes a multistep tunable SR network named MTSR, which simulates the mapping process from low-resolution inputs to high-resolution outputs and introduces adjustable parameters, and enables the method to produce nonsmooth SR results without artifacts while preserving rich and realistic textural details.
Guodong Ding, Jinhe Hu, Wei Xue et al.· IEEE Journal of Selected Top...· 0 citations
Remote sensing images captured under haze, mist, or thin-cloud conditions usually suffer from contrast attenuation, detail blurring, and spectral distortion, which reduces image interpretability and weakens the reliability of subsequent quantitative applications. Although CNN and Transformer-based restoration methods have achieved notable progress, CNN models are limited in global context perception, whereas Transformer architectures generally introduce considerable computational overhead when processing high-resolution remote sensing images. State-space models, especially Mamba, provide an efficient solution for long-sequence modeling with linear complexity. Nevertheless, existing Mamba-based restoration frameworks are still insufficient in jointly representing spatial structures, channel correlations, and degradation-adaptive feature responses. To address these limitations, we present TEMamba, a tri-scanning state-space model with multi-expert modulation for remote sensing image dehazing. The proposed network introduces a tri-scanning state-space block, which converts feature representations into complementary scanning sequences along horizontal, vertical, and channel-related directions. In this way, the model can better capture long-range spatial continuity, inter-channel dependency, and nonuniform haze distribution. Moreover, a multi-expert-driven aggregator is designed to dynamically integrate discriminative spatial and channel representations, enabling adaptive feature refinement under heterogeneous degradation conditions. In addition, a multidomain joint optimization objective is employed to constrain the reconstruction process from pixel, edge, and frequency perspectives, thereby improving structural preservation and spectral consistency. Experiments on representative remote sensing dehazing benchmarks demonstrate that the proposed method achieves competitive restoration performance compared with existing state-of-the-art approaches.
Unknown authors· Journal of Applied Remote Se...· 0 citations
Abstract. Satellite remote sensing is inherently constrained by a trade-off between spatial and temporal resolution. As a result, high-temporal-frequency sensors such as Geostationary Ocean Color Imager-II provide operationally valuable observations but at coarse spatial resolution. Reference-Based Super-Resolution (Ref-SR) can address this limitation by transferring high-resolution textures from an external reference image, but temporal mismatch between the target and reference images often leads to unreliable texture transfer and severe artifacts. This problem becomes more critical in extreme low-resolution (LR) settings, where structural information is already severely degraded. To address this issue, we propose the Dynamic Ref-SR Framework, which computes a pixel-wise weight map from intensity differences between the LR and reference images to selectively control reference transfer. The resulting weights promote reference use in stable regions while suppressing it in temporally inconsistent regions. The framework was validated on three backbone architectures—CNN (EDSR), Swin Transformer, and GAN—using a Sentinel-2 dataset for four-band reconstruction (RGB and NIR). Across all metrics and architectures, the proposed Ref-SR framework consistently outperformed the SISR baseline in both structural and spectral evaluations. Among the tested backbones, the GAN-based model achieved the best overall performance, with a PSNR of 35.60 dB, an SSIM of 0.92, a SAM of 2.20°, and an ERGAS of 74.71. These results demonstrate that the proposed framework can improve LR satellite imagery while reducing the risk of reference misuse under temporal mismatch.
Chae-Eun Kim, Junhwa Chi· The International Archives o...· 0 citations
Remote sensing image super-resolution is a critical task for reconstructing high-fidelity images from low-resolution observations. However, practical remote sensing scenarios often involve complex, compound degradations that severely compromise essential high-frequency structures, such as sharp edges and fine textures. While conventional convolutional neural network and Transformer-based methods have shown promise, they predominantly rely on static feature representations that lack adaptability to diverse scenes and often struggle to restore specific high-frequency details due to spectral bias. To this end, we propose a novel frequency-aware prompt learning framework. Specifically, a frequency-aware transformer module is designed to leverage the fast Fourier transform for explicit high-frequency component restoration, thereby overcoming the frequency bias inherent in spatial-domain models and effectively recovering intricate structural details. Concurrently, a dynamically generated prompt modulator is introduced to provide scene-specific adaptability via learnable vectors. This allows the network to adaptively calibrate feature responses to mitigate diverse environmental variations and compound artifacts. This synergistic integration ensures superior reconstruction fidelity and robust generalization across heterogeneous remote sensing scenarios. Comprehensive experiments across three benchmark datasets demonstrate that our method achieves outstanding performance in both quantitative metrics and visual quality assessments.
Kangli Zeng, Hang Hu, Ying Yu et al.· IEEE Journal of Selected Top...· 0 citations
Remote sensing images are frequently degraded by occlusions and missing observations, which significantly affect subsequent interpretation and analysis. Matrix completion provides an effective solution for recovering incomplete data; however, existing deep learning-based approaches often rely on random initialization, resulting in slow optimization and limited reconstruction quality under severe missing conditions. To address these issues, this paper proposes a two-stage neural network-based matrix completion framework that combines SVD-guided low-rank modeling with convolutional feature learning. Specifically, truncated singular value decomposition (SVD) is first employed to initialize the network and provide a coarse reconstruction by jointly modeling the global low-rank structure and nonlinear image representations. A U-Net-based convolutional autoencoder is then used to refine the reconstruction by exploiting local spatial correlations and multi-scale features. In addition, a channel aggregation strategy is introduced to improve structural consistency for multi-channel remote sensing images. The proposed framework adopts a training-data-free optimization paradigm, eliminating the need for external training datasets by optimizing the network parameters directly for each input image. Experimental results on synthetic and real remote sensing images demonstrate that the proposed method consistently outperforms conventional matrix completion methods and achieves competitive performance compared with recent deep learning approaches, particularly under random missing patterns and high missing-rate scenarios.
Jie He, Zijian Lin, Tianyao Huang et al.· Remote Sensing· 0 citations
Remote sensing image super-resolution (RSISR) provides an effective means of improving spatial detail for Earth observation and satellite image interpretation. However, existing methods often rely on increasingly complex network designs with deeper hierarchies and expanded channel capacities to pursue higher performance, resulting in heavy models with high computational cost, which restricts their deployment on resource-constrained platforms. To address this challenge, we propose a novel reparameterized feature enhancement network (RepFEN) for lightweight and accurate RSISR tasks. Specifically, a multi-scale reparameterized module (MRepM) is designed to capture multi-scale spatial information and enhance texture representation. Furthermore, a partial-channel gated attention module (PCGAM) is introduced to selectively enhance discriminative features along the channel dimension, effectively improving fine-grained detail restoration. By integrating structural reparameterization and multi-scale lightweight modules, the proposed method achieves a better balance between reconstruction accuracy and inference efficiency. Extensive experiments on both remote sensing and natural image super-resolution benchmarks demonstrate that our method achieves superior performance compared to existing state-of-the-art methods, while maintaining minimal computational overhead, showing significant potential for real-world applications.
