Multi-Resolution SAR and Optical Remote Sensing Image Registration Methods: A Review, Datasets, and Future Perspectives

2025-02-03

AI Summary

This work addresses three key challenges in multi-resolution SAR–optical remote sensing image registration: (1) mismatch of high-resolution structural details, (2) stereoscopic spatial misalignment, and (3) absence of standardized benchmark datasets. To this end, we present a systematic survey of existing methods, introduce MultiResSAR—the first publicly available multi-source, multi-resolution, multi-scene registration dataset comprising over 10,000 image pairs—and conduct a comprehensive evaluation of 16 state-of-the-art algorithms. Experimental results reveal a sharp degradation in registration accuracy with increasing resolution; all evaluated methods fail completely on sub-meter-resolution data. Among traditional methods, RIFT achieves the highest success rate (66.51%), while XoFTR leads among deep learning-based approaches (40.58%). Based on these findings, we propose three novel research directions: noise-robust feature suppression, 3D geometric fusion, and cross-view representation modeling—thereby establishing a foundational dataset, standardized evaluation protocol, and technical roadmap for high-resolution heterogeneous remote sensing image registration.

Technology Category

• Machine Learning: Multi-instance/Multi-view Learning
• Intelligent Robots: Multimodal Perception & Sensor Fusion
• Computer Vision: Multi-modal Vision

Application Category

• Search and Retrieval-Augmented AI: Web evaluation methodologies and metrics
• Systems and Infrastructure for Web, Mobile and WoT: Experiences and lessons learnt from Web-based algorithms and system deployments
• Graph Algorithms and Modeling for the Web: Algorithms and analysis for heterogeneous, signed, attributed, multi-relational, temporal, higher-order, and annotated Web-related graphs

Abstract

Synthetic Aperture Radar (SAR) and optical image registration is essential for remote sensing data fusion, with applications in military reconnaissance, environmental monitoring, and disaster management. However, challenges arise from differences in imaging mechanisms, geometric distortions, and radiometric properties between SAR and optical images. As image resolution increases, fine SAR textures become more significant, leading to alignment issues and 3D spatial discrepancies. Two major gaps exist: the lack of a publicly available multi-resolution, multi-scene registration dataset and the absence of systematic analysis of current methods. To address this, the MultiResSAR dataset was created, containing over 10k pairs of multi-source, multi-resolution, and multi-scene SAR and optical images. Sixteen state-of-the-art algorithms were tested. Results show no algorithm achieves 100% success, and performance decreases as resolution increases, with most failing on sub-meter data. XoFTR performs best among deep learning methods (40.58%), while RIFT performs best among traditional methods (66.51%). Future research should focus on noise suppression, 3D geometric fusion, cross-view transformation modeling, and deep learning optimization for robust registration of high-resolution SAR and optical images. The dataset is available at https://github.com/betterlll/Multi-Resolution-SAR-dataset-.

Problem

• SAR-Optical Image Alignment
• High-Resolution Imagery
• Multi-Resolution Dataset

Innovation

• MultiResSAR Dataset
• High-resolution Image Alignment
• Performance Evaluation of Algorithms