Neptune: An Automated System for Dark Ship Detection, Targeting, and Prioritization
Abstract
Dark ship detection at open ocean scale drives the need for enhanced space-based intelligence, surveillance, and reconnaissance capabilities. With the boom of commercial space-based sensing, an automated process is needed to meet the growing volume and velocity of data. Aggregation and fusion of multi-modal data from the variety of existing and proposed space-based sensor networks can be leveraged to produce target quality tracks on ships. These sensor modalities include SAR (Synthetic Aperture Radar), EO/IR (Electro-optical / Infrared), and AIS (Automatic Information System). In this paper we demonstrate the ability to perform automated target recognition of surface vessels from these modalities on a space-flight processor to simulate on-orbit detection. These detections are fused to form quality tracks which can then be used for anomaly detection of dark ships via pattern of life. Tracks formed over continental or global scale motivates the need for further automated analysis. A significant amount of human effort would be needed to analyze thousands or tens of thousands of tracks in detail and in real-time. We developed a suite of pattern-of-life tools that extracts features from tracks and flags tracks that deviate too far from some learned definition of normality.
Document Details
- Document Type
- Technical Report
- Publication Date
- Dec 14, 2021
- Accession Number
- AD1173758
Entities
People
- Adam Byerly
- Musad Haque
- Sesan Iwarere
- Sheldon Bish
- Tamim Sookoor
- Waseem Malik
- Will Zhang
Organizations
- Johns Hopkins University Applied Physics Laboratory