Associate Researcher and Master's Supervisor at Wuhan University. The main research directions include intelligent interpretation of remote sensing images, image semantic segmentation and change detection based on deep learning, design of remote sensing image sample libraries and specialized deep learning framework models, design of remote sensing large models, and multi-agent driven extraction of remote sensing land features.

Hosted or participated in more than 20 horizontal and vertical projects such as the National Natural Science Foundation and Key R&D Programs, and has published more than 20 academic papers. Served as a reviewer for top computer vision and pattern recognition conferences such as CVPR, ICCV, ECCV, ISPRS J P&RS, and remote sensing journals for a long time. Received awards such as "Special Prize in Surveying and Mapping Science and Technology" (Rank 2), "Special Prize in Geographic Information Technology Progress" (Rank 3), and "First Prize in Xia Jianbai Science and Technology Innovation". The automatic/semi-automatic interpretation system EasyFeature, which was developed as a core personnel, has been used in major national engineering projects such as "Global Mapping" and "Geographical Survey". It has supported the production of 1:10000/50000 global digital maps of over 20 million square kilometers in Asia, Africa, Europe, North America, and other regions. The collection efficiency has increased by about 50%, and more than 1000 sets have been promoted in the industry, with economic benefits exceeding 200 million yuan. The development of the world's first remote sensing deep learning framework, LuoJiaNET, with expandable memory and flexible channel selection, has achieved full stack autonomous control of remote sensing feature frameworks, attracting more than 5000 users from more than 20 countries and regions in Europe, the United States, and Asia, providing a basic foundation for autonomous control of natural resource monitoring, open earth engine OGE, and other projects. The achievement has won the Hubei Artificial Intelligence Major Technology Innovation Achievement Award and has been reported by media such as People's Daily and Changjiang Daily. At the same time, the LuoJiaSET classification system of "scene target pixel" was proposed, which supported the establishment of the OGC Training DML classification system by the Geographic Information Standardization Organization. The relevant sample set was selected by the National Natural Science Foundation of China's Intelligent Remote Sensing Interpretation Competition, attracting more than 700 domestic and foreign participants and more than 440 participating teams.