PhyDAE: Physics-Guided Degradation-Adaptive Experts for All-in-One Remote Sensing Image Restoration

2025-10-09

AI Summary

Remote sensing images are often degraded by heterogeneous factors—including atmospheric interference, noise, blur, and low illumination—posing challenges for unified restoration. Existing implicit modeling approaches suffer from poor interpretability and limited generalization. To address this, we propose the Physics-guided Degradation-adaptive Expert Network (PAEN), the first framework to explicitly map implicit features into degradation decision signals. PAEN introduces a Residual Manifold Projector (RMP) to model geometric structural degradation and a Frequency-domain Degradation Decomposer (FADD) to disentangle multi-scale degradation components. It further incorporates physics-aware expert modules with a temperature-controlled sparse activation mechanism. This two-stage cascaded architecture tightly integrates manifold geometry and frequency-domain priors. Evaluated on three benchmark datasets, PAEN consistently outperforms state-of-the-art methods, achieving significant PSNR/SSIM gains while reducing parameter count and FLOPs by over 40%, thereby unifying high-fidelity restoration with efficient inference.

Technology Category

• Computer Vision: Low Level & Physics-based Vision
• Machine Learning: Deep Generative Models & Autoencoders
• Intelligent Robots: Multimodal Perception & Sensor Fusion

Application Category

• Search and Retrieval-Augmented AI: Web learning to rank, online learning, and counterfactual learning for ranking
• Graph Algorithms and Modeling for the Web: Graph neural networks and deep learning approaches for Web-related graphs
• Responsible Web: Machine-in-the-loop, human agency and autonomy

Abstract

Remote sensing images inevitably suffer from various degradation factors during acquisition, including atmospheric interference, sensor limitations, and imaging conditions. These complex and heterogeneous degradations pose severe challenges to image quality and downstream interpretation tasks. Addressing limitations of existing all-in-one restoration methods that overly rely on implicit feature representations and lack explicit modeling of degradation physics, this paper proposes Physics-Guided Degradation-Adaptive Experts (PhyDAE). The method employs a two-stage cascaded architecture transforming degradation information from implicit features into explicit decision signals, enabling precise identification and differentiated processing of multiple heterogeneous degradations including haze, noise, blur, and low-light conditions. The model incorporates progressive degradation mining and exploitation mechanisms, where the Residual Manifold Projector (RMP) and Frequency-Aware Degradation Decomposer (FADD) comprehensively analyze degradation characteristics from manifold geometry and frequency perspectives. Physics-aware expert modules and temperature-controlled sparse activation strategies are introduced to enhance computational efficiency while ensuring imaging physics consistency. Extensive experiments on three benchmark datasets (MD-RSID, MD-RRSHID, and MDRS-Landsat) demonstrate that PhyDAE achieves superior performance across all four restoration tasks, comprehensively outperforming state-of-the-art methods. Notably, PhyDAE substantially improves restoration quality while achieving significant reductions in parameter count and computational complexity, resulting in remarkable efficiency gains compared to mainstream approaches and achieving optimal balance between performance and efficiency. Code is available at https://github.com/HIT-SIRS/PhyDAE.

Problem

• Addressing multiple heterogeneous degradations in remote sensing images
• Overcoming limitations of implicit feature representations in restoration methods
• Improving restoration quality while reducing computational complexity

Innovation

• Two-stage cascaded architecture transforms implicit features into explicit decisions
• Progressive degradation mining analyzes manifold geometry and frequency characteristics
• Physics-aware expert modules with sparse activation ensure computational efficiency