Học Liên Kết Để Phát Hiện Đối Tượng Tàu: Khảo Sát Các Phương Pháp, Thách Thức Và Hướng Đi Trong Tương Lai
Year: 2026
Type: article
Abstract: Maritime domains are critical to economic development and international trade, while also playing a key role in the protection of national sovereignty. The exploitation of marine resources relies heavily on maritime vessels, which are central to transportation and offshore operations. For coastal regions frequently affected by natural disasters such as storms and tsunamis, there is a pressing need for reliable ship detection systems using aerial drones, satellites, radar, or surveillance cameras. Such systems enable timely identification and coordination of vessels at sea, facilitating evacuation to safe zones during extreme events. In addition, ship detection is essential for monitoring unauthorized maritime activities and illegal incursions into territorial waters. While recent studies have explored the application of Machine Learning, Deep Learning, and Federated Learning in maritime environments, a gap remains in integrating these approaches into unified frameworks. This work aims to review and analyze these methodologies for ship detection, with the goal of identifying suitable algorithms and models for practical deployment under diverse operational conditions.
Language: English
Subfield: Computer Vision and Pattern Recognition
Field: Computer Science
Domain: Physical Sciences
Open Access status: diamond
OpenAlex ID: W7220132564