Year: 2025
Type: conference-paper
Abstract: Marine target recognition technology from Synthetic Aperture Radar (SAR) images is of great significance in both civil and military fields, so it has become a research hotspot in various countries. However, in complex sea scenes, the existing Marine target detection and recognition technologies have poor accuracy and robustness in high-complexity and strong interference scenes, and cannot stably identify Marine targets in the face of various interference factors. This paper focuses on image preprocessing, and proposes a selection algorithm for the region of interest of the sea surface and a clustering and enhancement algorithm for potential targets based on sea-land separation. This method clusters strong points of the ship SAR image into a single connected surface target, and verifies that it can lay a good foundation for subsequent detection and recognition.
Authors: Boxuan Liu, Zhongma Cui
Language: English
Subfield: Aerospace Engineering
Field: Engineering
Domain: Physical Sciences
Open Access status: hybrid
OpenAlex ID: W4411539893