HFGNet: Mitigating Boundary Distortion for Sonar Image Segmentation with High Frequency Guidance Strategy

Abstract

Sonar image segmentation often suffers from blurred or distorted target boundaries. HFGNet addresses this problem with a high-frequency guidance strategy, using boundary-sensitive information to mitigate segmentation distortion and improve target delineation in sonar imagery.

Publication
In 2026 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP)
Shaojian Yang
Shaojian Yang
Assistant Research Fellow / Postdoctoral Fellow

My research interests include underwater communications, distributed acoustic sensing and ocean Internet of Things.