Search Results (3)

Ship type (e.g., Cargo, Tanker and Fishing) classification is crucial for marine management, environmental protection, and maritime safety, as it enhances navigation safety and aids regulatory agencies in combating illegal activities. Traditional ship type classification methods with AIS data are often plagued by problems such as data imbalance, insufficient feature extraction, reliance on single-model approaches, or unscientific model combination methods, which reduce the accuracy of classification. In this paper, we propose an ensemble classification method based on a stacking strategy to overcome these challenges. We apply the SMOTE technique to balance the dataset by generating minority class samples. Then, a more comprehensive ship behavior model is developed by combining static and dynamic features. A stacking strategy is adopted for the classification, integrating multiple tree structure-based classifiers to improve classification performance. The experimental results show that the ensemble classification method based on the stacking strategy outperforms traditional classifiers such as CatBoost, Random Forest, Decision Tree, LightGBM, and the ensemble classification method, especially in terms of improving classification precision, recall, F1 score, ROC curve, and AUC. This method improves the accuracy of ship type recognition, and it is suitable to real-time online classification, which is helpful for applications in marine safety monitoring, law enforcement, and illegal fishing detection. Full article

The rising tensions in disputed waters in Southeast Asia have caused policymakers to diverge their maritime diplomatic strategy to include maritime constabulary forces. The use of coastguards and other non-military vessels are an emerging trend in the maritime diplomatic strategy of Southeast Asian states, including in the high-profile case of the North Natuna Seas, to which scholars pay little attention. This article contends that (1) contemporary maritime diplomacy in Southeast Asia positions the utilization of maritime constabulary forces (coastguards, maritime law enforcement agencies) as its primary maritime diplomatic strategy; (2) Vietnam’s coercive turn in its maritime disputed areas was a deliberate attempt to balance a coercive-cooperative stance against Indonesia in the North Natuna Seas, following its traditional coercive maritime diplomatic stance against China, and; (3) Vietnam’s utilization of maritime constabulary forces as a measure to solidify its sovereign claims coincided with the benefits of tactical military flexibility and non-escalatory means to achieve its aims in the Natuna Seas. This empirical explanatory research delves into the development of Vietnam’s coastguards and maritime law enforcement agencies by interpreting the secondary data from the Indonesia Ocean Justice Initiative between 2021–2022 on cases relating to suspicious maneuvers conducted by the Vietnamese Fisheries Resource Surveillance vessels safeguarding the conduct of Vietnamese IUUF. Full article

As commercial geospatial intelligence data becomes more widely available, algorithms using artificial intelligence need to be created to analyze it. Maritime traffic is annually increasing in volume, and with it the number of anomalous events that might be of interest to law enforcement agencies, governments, and militaries. This work proposes a data fusion pipeline that uses a mixture of artificial intelligence and traditional algorithms to identify ships at sea and classify their behavior. A fusion process of visual spectrum satellite imagery and automatic identification system (AIS) data was used to identify ships. Further, this fused data was further integrated with additional information about the ship’s environment to help classify each ship’s behavior to a meaningful degree. This type of contextual information included things such as exclusive economic zone boundaries, locations of pipelines and undersea cables, and the local weather. Behaviors such as illegal fishing, trans-shipment, and spoofing are identified by the framework using freely or cheaply accessible data from places such as Google Earth, the United States Coast Guard, etc. The pipeline is the first of its kind to go beyond the typical ship identification process to help aid analysts in identifying tangible behaviors and reducing the human workload. Full article