Optical and SAR Image Fusion: A Review of Theories, Methods, and Applications

Abstract

Highlights: What are the main findings? This review systematically summarizes the theoretical foundations of optical and synthetic aperture radar (SAR) imaging, clarifying their complementarity for remote-sensing image interpretation. It highlights the key role of optical and SAR data fusion (OPT-SAR fusion) while outlining the evolution of related methods, open data and code resources, and emerging development trends. What are the implications of the main findings? This review enables readers to quickly and comprehensively grasp the current status of OPT-SAR fusion methods and their applications in relevant Earth observation missions from a technical perspective. By consolidating future research directions in OPT-SAR fusion, this review outlines a technical roadmap toward more reliable, generalizable, and practically deployable fusion systems. Remote sensing technology has become an indispensable core means for Earth observation. As two of the most commonly used remote sensing modalities, the fusion of optical and synthetic aperture radar (SAR) (OPT-SAR fusion) can effectively overcome the limitations of a single data source, achieve information complementarity and synergistic enhancement, thereby significantly improving the interpretation capability of multi-source remote sensing data. This paper first discusses the necessity of OPT-SAR fusion, systematically reviews the historical development of fusion technologies, and summarizes open-source resources for various tasks, aiming to provide a reference for related research. Finally, building upon recent advances in OPT-SAR fusion research and cutting-edge developments in deep learning, this paper proposes that future fusion technologies should develop in the following directions: interpretable fusion models driven by both data and knowledge, general fusion perception driven by multimodal large models, and lightweight architectures with efficient deployment strategies.