He focuses on developing high-quality precipitation datasets by integrating radar and satellite observations with advanced machine learning techniques. He aims to improve the accuracy and reliability of rainfall estimation and prediction through the effective fusion of multi-source remote sensing data and data-driven models. His research particularly addresses the limitations of individual observation systems, such as spatial gaps, measurement uncertainty, and differences in temporal and spatial resolution.

​

He also extends his research to flood detection and the prediction of inundation-prone areas under extreme rainfall conditions. By linking high-resolution precipitation information with hydrological analysis and disaster-related applications, he seeks to support more effective risk assessment, early warning, and emergency response. Ultimately, he hopes to contribute to reducing human casualties and societal damage caused by climate change–induced extreme rainfall events through AI-based and data-informed approaches.