Year: 2025
Type: article
Abstract: With the rapid development of the marine economy, maritime traffic supervision and safety management have raised higher requirements for ship detection. Traditional methods suffer from low efficiency and high false detection rates, making it difficult to meet the needs of intelligent maritime management. This paper proposes a ship target detection model based on YOLOv8. Experimental results show that the model achieves 99.15% mAP@50 and 85.14% mAP@50:95, effectively handling ship recognition tasks under complex sea conditions, providing reliable technical support for smart ocean construction and maritime supervision.
Author: Wanqiu Xu
Institution: Jiangsu Maritime Institute
Language: English
Subfield: Computer Vision and Pattern Recognition
Field: Computer Science
Domain: Physical Sciences
Open Access status: bronze
OpenAlex ID: W4414623019