Answers are grounded in this paper and cite their sources.
Remote sensing images (RSIs) object detection is a product of the deep integration of computer vision and Earth observation technology, and its development has significantly enhanced humanity's ability to perceive and interpret surface information. This review systematically examines the research progress in this field from three dimensions: data, technology, and applications. At the data level, we explore the evolution from scarce to abundant datasets and the associated annotation techniques. At the technological level, we focus on the evolution of detection frameworks, ranging from handcrafted feature-based methods to modern deep learning paradigms, as well as specialized architectures designed to address challenges such as multi-scale objects and large-format images in remote sensing. From the application perspective, this paper adopts a dual viewpoint of platforms and disciplines, elaborating on the typical applications of satellite imagery in large-scale macroscopic monitoring and UAV imagery in fine-grained real-time recognition. It also discusses how remote sensing object detection technology serves as a core tool, deeply enabling cross-disciplinary integration and paradigm innovation in fields such as geography, ecology, and urban planning. Finally, this paper summarizes the existing gaps between theoretical models and practical applications and offers prospects for future advancements toward more intelligent and integrated cross-disciplinary solutions.
No takes yet. Share an insight, caveat, or question.
Cheng et al. (2026) studied this question.
Synapse has enriched 10 related papers. Consider them for comparative context: