SAR Vessel Detection and Gross Tonnage Estimation from Heterogeneous Datasets for Dark Vessel Identification

2026-07-20

AI Summary

This study addresses the challenge of identifying “dark vessels”—ships that disable transponders to evade regulatory oversight—by simultaneously detecting vessels in synthetic aperture radar (SAR) imagery and estimating their gross tonnage (GT) to determine whether they are legally required to carry a transponder. To this end, the authors propose the first end-to-end multitask deep learning framework that fuses heterogeneous SAR images with vessel attribute data to jointly predict vessel location, type, and physical dimensions. A cascaded non-parametric KNN regression model, incorporating a hybrid distance metric that combines Euclidean distance with category semantics, is introduced to achieve high-accuracy GT estimation. The method achieves state-of-the-art performance on both vessel detection and GT regression subtasks, effectively enabling automated dark vessel identification, and the implementation code has been publicly released.

Abstract

Detecting vessels engaging in illegal activities is of paramount importance for maritime security. One of the major goals is to detect dark vessels, ships that disable their transponders to evade surveillance. Deep Learning (DL) models can detect vessels in Synthetic Aperture Radar (SAR) images, enabling maritime traffic analysis regardless of weather or visibility conditions. However, to detect potential dark vessels, a DL model must select only those that are required to carry a transponder based on their Gross Tonnage (GT). Unfortunately, no public SAR dataset is available for training an end-to-end DL model for vessel detection and GT regression. In this work, we present a framework that leverages heterogeneous image and tabular datasets to solve this task. Our solution combines a multi-task DL framework for predicting the location, vessel type, and physical dimensions of ships, cascaded with a non-parametric model for predicting GT from vessel size and category. We perform GT regression by a KNN that measures sample similarity using a hybrid Euclidean and categorical distance. Experiments show that our solution can predict multiple outputs while remaining competitive with state-of-the-art models on individual subtasks, thus enabling the identification of dark vessels. We publish our code on GitHub https://github.com/PaltrinieriDavide/vesseldetection.

Problem

• SAR vessel detection
• Gross Tonnage estimation
• dark vessel identification
• heterogeneous datasets
• maritime surveillance

Innovation

• multi-task deep learning
• heterogeneous data fusion
• gross tonnage estimation
• dark vessel detection
• non-parametric regression