Maritime Analytics Prototype: Final Development Report
Abstract
This report documents the implementation of a Maritime Visual Analytics Prototype (MVAP), a test-bed and showcase for novel visual analytics strategies that will support maritime situation awareness, anomaly detection, and detailed analysis of a Vessel of Interest (VOI). MVAP uses a Service Oriented Architecture, with reuse of services from DRDC Valcartiers Intelligence Science and Technology Integration Platform (ISTIP).The development took place in Phases from 2012 to 2014. In Phase 1, four novel widgets or apps were developed. The Vessel Summary Cards deliver a unique visual representation of each ship, suitable for rapid visual browsing and data drill-down. The Record Browser widget allows analysts to rapidly shuffle through the Vessel Summary Cards to gain a visual impression of their assigned set of vessels without having to read all the details. The Timeline widget can show timelines of multiple ships, offset if necessary to reveal recurrent patterns of behavior. The Magnets Grid widget uses a dynamic scatterplot to sort events and objects based on their multi-dimensional properties. These widgets are supported by an Analysis Set Manager service that allows users to create and modify groups of ships and pass those groups between widgets. Additionally, work was undertaken to analyze and improve upon the underlying web development framework known as the Widget Application Shell (WAS). In Phase 2, the Map and Timeline widget was developed to allow the user to interactively view time-varying vessel positional data and vessel contact data. Capabilities developed in Phase 3 included extending MVAP to improve the ability to deploy the visualizations as part of a shared collaborative workspace, and integrating artificial reasoning capabilities to manage smart sets of vessels. As part of the Network Analysis Capabilities amendment MVAP widgets were modified to support network graph analysis, and a new Graph Analyzer widget was developed.
Document Details
- Document Type
- Technical Report
- Publication Date
- Apr 01, 2014
- Accession Number
- AD1003620
Entities
People
- Eric J Hall
- Michael L. Davenport
- Neil Bozowsky
- William Wright
Organizations
- Oculus Info Inc.