Prof. Kegen Yu, IEEE Senior Member

China University of Mining and Technology, China

Prof. Kegen Yu received the PhD degree in electrical engineering from The University of Sydney, Australia, in 2003. He has worked for universities and research institutions in Australia, China and Finland. He is currently a Distinguished Professor with China University of Mining and Technology, Xuzhou, China. Prof. Yu has co-authored the book "Ground-Based Wireless Positioning" (Wiley-IEEE Press, 2009) and another book titled "Wireless Positioning: Principles and Practice" (Springer, 2018). Additionally, he authored the book "Theory and Practice of GNSS Reflectometry" (Springer, 2021). He has contributed to over 200 refereed journal and conference articles, including more than 70 articles published in IEEE journals. He was ranked in the world’s top 2% scientists list in 2022 by Stanford University and Elsevier. His research interests include GNSS-R, wireless positioning, and remote sensing.

Title: Measuring Soil Moisture with GNSS Reflectometry
Abstract: As an emerging remote sensing technology, GNSS reflectometry (GNSS-R) has been employed for retrieving environmental and geophysical parameters and monitoring land and ocean disasters. In this presentation, we will talk about some fundamental concepts and principles of this technology. Recent satellite missions associated with GNSS-R will first be briefly introduced. We will then study how to use ground-based and spaceborne GNSS-R data to measure soil moisture. Retrieval modeling based on curve fitting and machine learning will be studied. Some experimental results will also be presented.

Prof. Yong Wang

University of Electronic Science and Technology of China, China

Yong Wang received the Ph.D. degree in computer science and technology from the Institute of Computing Technology, Chinese Academy of Sciences, Beijing, China, in 2008. He is currently an Associate Professor with the School of Computer Science and Engineering, University of Electronic Science and Technology of China. His research interests include spatial database, spatial query processing, and privacy enhancing technologies.

Title: BiSAR Data Analysis: An Approximation by Surface Scattering Modeling
Abstract: Bistate Synthetic Aperture Radar (BiSAR) enhances Earth observation efficiency by simultaneously receiving radar echoes from two receivers on primary and secondary platforms. When configured cross-track, BiSAR reduces temporal decorrelation, enabling precise topography measurement. For topography-focused BiSAR, the small separation distance between the two platforms results in minimal bistate angles, allowing the primary and secondary platform-to-target distances to be approximately equal. Then, the use of a backscatter-based approach for BiSAR data analysis is argued. By modeling the backscattering and bistate scattering from the surface and considering noise equivalent sigma zero (NESZ), the validity region of the approximation within bistate scattering angles has been determined. For smooth or slightly rough surfaces modeled by the small perturbation model, the region ranged from 0° to 0.056° (L-HH and an incidence angle of 60°) and from 0° to 0.137° (L-VV and an incidence angle of 20°). The region for rough surfaces, as predicted by the geometric optical model, ranged from 0.028° to 0.634°. Thus, the argument of using the approximation has been confirmed. More importantly, the BiSAR scattering analysis has been simplified, and the understanding of the obtained BiSAR data has been furthered.

Prof. Mahmoud Reza Delavar

University of Tehran, Iran

Prof. Mahmoud Reza Delavar is Director, Center of Excellence in Geomatic Eng. in Disaster Management and Director, Land Administration in Smart City Lab., School of Surveying and Geospatial Eng., College of Engineering, University of Tehran. Dr. Delavar has received his BSc. in Civil Eng.-Surveying from Technical University of KNT, Tehran, Iran in 1989, MSc in Civil Eng. - Photogrammetry and Remote Sensing from University of Roorkee (Currently IIT Roorkee), Roorkee, India in 1992 and a PhD in Geomatic Eng.-GIS from University of New South Wales (UNSW), Sydney, Australia in 1997. He was the chair of International Society of Photogrammetry and Remote Sensing ISPRS WG IV/3 (Spatial Statistics, Analysis and Uncertainty Modeling) during 2016- 2022, chair of ISPRS WG IV/2 (Artificial Intelligence and Uncertainty Modeling in Spatial Analysis) during 2022-2024 and is the advisor of the same working group during 2024-2026. He is Iran's national representative to International Society of Urban Data Management (UDMS) since 2006. Prof. Delavar is founder of Iranian Society of Surveying and Geomatics Eng. (ISSGE) and is in the editorial board of ISPRS International Journal of Geo-Information (IJGI) and Geo-spatial Information Science (GSIS). Prof. Delavar is a member of International Society of Urban Informatics. Prof. Delavar has published 404 papers in reputable national and international conferences and Journals. Prof. Delavar has supervised 124 MSc. and PhD theses and Postdoc research so far. His research interests are in spatial data quality and uncertainty modeling, temporal GIS, disaster management, smart cities, cadaster, land administration, spatial data infrastructure (SDI), building information modeling/management, (BIM), multi-dimensional GIS, ubiquitous GIS, spatial interoperability, spatial data fusion, spatial data science, intelligent GIS, urban growth modeling, land use and land cover change modeling, and integration of remote sensing and GIS.

Prof. Zhi Gao, IEEE Member

Wuhan University, China

Gao Zhi, Professor and doctoral supervisor at the School of Remote Sensing and Information Engineering, Wuhan University, and Vice Dean of the Wuhan University College of Excellent Engineers, previously served as Vice Dean of the School of Remote Sensing, Deputy Director of the International Exchange Office at Wuhan University, recipient of the National High-Level Overseas Talents Program (Youth Project), Distinguished Professor of Chu Tian Scholars in Hubei Province, Outstanding Young Scientist in Hubei Province, and leader of major projects under the National Natural Science Foundation of China. Professor Gao Zhi's research focuses include remote sensing big data and intelligent interpretation of remote sensing images, computer vision, and intelligent unmanned systems. He has led 7 government-funded research projects in Singapore, 1 major project under the National Natural Science Foundation of China, 8 provincial and ministerial-level projects, and over 20 projects for multinational corporations. He has published more than 100 academic papers, including 68 SCI papers in top journals such as IEEE PAMI, IJCV, IEEE TGRS, and ISPRS J Photogram Remote Sensing, as well as 25 conference papers recommended by the China Computer Federation for Class A and B. He serves as a specially appointed expert for the China Association for Science and Technology's HaiZhi Program, editorial board member of 3 journals, and has been a program committee member and sub-forum chair at multiple international conferences, as well as deputy secretary-general of a specialized committee.

Title: Research on Real-Time Remote Sensing of Surface Anomalies
Abstract: Over the past few decades, abnormal phenomena on Earth's surface—including natural disasters and human-induced events—have become increasingly frequent, resulting in massive casualties, population displacement, and property losses. Timely anomaly detection and response have therefore become critical. To achieve rapid detection, we propose a hierarchical anomaly detection framework based on Graph Neural Networks (GNNs), termed L2S-Net, designed to integrate information from local to semantic levels (Local-to-Semantic, L2S). Specifically, L2S-Net utilizes a single ultra-high-resolution (VHR) remote sensing image as input to enhance processing speed. Inspired by brain-inspired research and graph theory, we developed a "local-to-semantic" fusion network—L2S-GNN—to explicitly learn relationships between nodes at different levels, thereby enabling more precise detection. Additionally, recognizing the crucial role of prior knowledge in anomaly classification, we embedded common object relationships from remote sensing scenes into the training and inference processes of L2S-GNN. Extensive experiments using the large-scale benchmark dataset ESADv2 and two real-world cases demonstrate that L2S-Net outperforms multiple state-of-the-art methods in accuracy while exhibiting excellent generalization capabilities and robustness.