Special Session 19  会议特别专题 19

Mechanisms, Models, and Applications of Interpretable Methods for Underwater Target Detection and Resolution

Description: While deep learning has achieved remarkable success in underwater target detection and resolution, model complexity and uninterpretability remain critical bottlenecks for deployment. The rise of interpretable methods brings a new perspective for understanding model mechanisms and enhancing model performance.
Therefore, this special session aims to explore the latest breakthroughs in interpretable theories, methods and models for underwater target detection and resolution. In this session, we focus not only on detection accuracy but also on why and how a model can detect a small and weak target from complex backgrounds, and whether its decision align with physical principles. The session will cover topics including discriminability mechanism, intrinsically interpretable models, post-hoc explanation algorithms, physics-informed data fusion, and uncertainty quantification. We seek contributions that bridge the gap between high performance and transparency, paving the way for robust intelligent systems in critical applications.

Session organizers
Assoc. Prof. Zipeng Li, Northwestern Polytechnical University, China
Prof. Hui Li, Northwestern Polytechnical University, China

The topics of interest include, but are not limited to:
▪ The Discriminability Mechanism of Active and Passive Sonar
▪ Inherently Interpretable Model for underwater target detection and resolution
▪ Post-hoc Explanation and Attribution Analysis for Detection and Resolution Tasks
▪ Trustworthy Decisions and Uncertainty Estimation for Multi-modal Fusion
▪ Applications of Explainable Methods in Engineering Practice

Submission method
Submit your Full Paper (no less than 5 pages with two colums) or your paper abstract-without publication (200-400 words) via Online Submission System, then choose Special Session 19 (Mechanisms, Models, and Applications of Interpretable Methods for Underwater Target Detection and Resolution)
Template Download

Introduction of Session organizer

Assoc. Prof. Zipeng Li, Northwestern Polytechnical University, China

Dr. Zipeng Li is an Associate Professor and Master’s Supervisor at the Ocean Institute, Northwestern Polytechnical University (NPU). Dr. Li commenced his studies at the School of Mechanical Engineering, Xi’an Jiaotong University (XJTU) in 2010. He earned dual bachelor’s degrees in Mechanical Engineering and Business Administration in 2014 and was subsequently recommended for direct admission to the doctoral program at XJTU. In 2017, sponsored by the China Scholarship Council (CSC), he conducted joint doctoral research at the University of Toronto, Canada. He was supervised by Prof. Viliam Makis, a Senior Member of both the Institute of Industrial and Systems Engineers (IISE) and the American Society for Quality (ASQ). Dr. Li obtained his Ph.D. in Mechanical Engineering from XJTU in 2020. Immediately thereafter, he joined NPU as a Postdoctoral Fellow, working in collaboration with Prof. Kunde Yang.
Dr. Li’s long-term research interests lie in adaptive signal processing, feature extraction, and pattern recognition. His work specifically addresses underwater acoustic signal processing, intelligent target detection and resolution and sonar technology.



Prof. Hui Li, Northwestern Polytechnical University, China

Dr. Hui Li is a Professor and Doctor’s Supervisor at the School of Marine Science and Technology, Northwestern Polytechnical University (NPU). Dr. Li received the B.S. degree in ocean technology from Dalian Ocean University (DLOU), Dalian, China, in 2012, and the M.S. and Ph.D. degrees in underwater acoustic engineering from Northwestern Polytechnical University (NPU), Xi’an, China, in 2013 and 2018, respectively. Since 2018, he has been with the School of Marine Science and Technology, NPU, where he is currently a professor. Dr. Li’s long-term research interests lie in deep ocean sound propagation, array signal processing, and source localization.