Bibliometric Analysis of the Research Progress on Cloud Removal of Optical Remote Sensing Images

High-resolution optical remote sensing images are widely used in environmental monitoring, disaster assessment, urban planning, and land use, but cloud cover seriously affects image quality and its application effect. In recent years, as artificial intelligence technology has advanced, cloud removal methods for optical remote sensing images have gradually evolved from machine learning to deep learning, but the development context and research hotspots of this field still lack systematic review. Based on the Web of Science Core Collection database, this paper uses the bibliometric analysis software VOSviewer to conduct a bibliometric analysis of 1,351 related literatures collected from 1999 to 2025, and systematically analyzes the development characteristics of the field of cloud removal of optical remote sensing images from the aspects of publication trend, country, institution, journal, author, and keywords. The results show that the research on cloud removal of optical remote sensing images has entered a rapid development stage since 2016, and the number of publications and citation frequency continue to grow; China and the United States are the most important research forces in this field, and institutions such as the Chinese Academy of Sciences, NASA, and Wuhan University, as well as core journals such as Remote Sensing and IEEE Transactions on Geoscience and Remote Sensing, have played an important role in this field; the keyword co-occurrence analysis identifies three main research directions: remote sensing applications and multi-source data fusion, cloud removal models and image reconstruction, and SAR-assisted cloud removal and multi-source image fusion.