Critical maritime areas require continuous surveillance and protection, relying on advanced military equipment of strategic importance. The interpretation of images obtained from these systems is essential for situational awareness and decision-making. Moreover, maritime logistics has become a cornerstone of global trade, making the classification and differentiation of ship types crucial for optimizing transportation efficiency, reducing storage costs, and enhancing security. This study focuses on the classification of ships engaged in various maritime missions, with an emphasis on military vessel detection and identification. To achieve high-accuracy ship classification, a deep learning-based approach was adopted. A comprehensive dataset of ship images was constructed using web scraping techniques from publicly available sources. Deep learning was preferred over traditional machine learning techniques due to its ability to extract high-level semantic features and learn complex patterns more effectively. The deep learning models were trained and evaluated on this dataset to optimize classification performance. Experimental results demonstrated classification accuracies ranging from 94% to 99%, highlighting the effectiveness of the proposed approach. This study presents the scientific findings, discusses the implications of the results, and explores potential applications in maritime surveillance and security.
Kaplan et al. (Tue,) studied this question.