Why activity-based intelligence and machine learning demonstrate that the future of GEOINT has already arrived
Editor’s Note: Barry Barlow is chief technology officer at Vencore. Guest posts are intended to foster discussion and do not represent the official position of USGIF or trajectory magazine.
GEOINT, shorthand for geospatial intelligence, is a term created by then director of the National Geospatial-Intelligence Agency (NGA), retired Air Force Lt. Gen. James R. Clapper, to define a fundamentally unique foundational element of intelligence. In his October 2005 memo on GEOINT, Clapper explained the subject as follows:
“GEOINT encompasses all aspects of imagery … and geospatial information and services. … It includes, but is not limited to … the analysis of literal imagery; geospatial data; and information technically derived from the processing, exploitation, literal, and non-literal analysis of spectral, spatial, and temporal … fused products (that is, products created out of two or more data sources). … These types of data can be collected on stationary and moving targets by electro-optical …, [synthetic aperture radar (SAR)] …, related sensor programs …, and non-technical means (to include geospatial information acquired by personnel in the field).”
A few future concepts from the abbreviated definition illustrate that the future of GEOINT is now.
Activity-Based Intelligence (ABI): This methodology to discover and resolve unknown entities and objects and depict a pattern of life is almost always included as a “future GEOINT trend” in any report on the subject. Clapper assumed ABI would be a foundational element of GEOINT—“these types of data can be collected on stationary and moving targets”—hence his desire to provide the broadest possible definition of GEOINT—“all aspects of imagery and geospatial information and services.” The incorporation of full motion video into the GEOINT domain was a leading indicator in the shift from reconnaissance to surveillance and from periodic collections to persistence, which is made easier by other future trends such as small satellites and the Internet of Things.
Two ABI precepts, data neutrality and integration before exploitation, highlight the need for another future trend: big data analytics. Big data analytics can provide hypotheses on the intention, strategies, or motivations of an adversary or ally. Ideally, analytics are anticipatory in nature and will be completed long before an issue appears on one’s radar as actionable intelligence. Again, Clapper’s definition of GEOINT was purposefully broad relative to data sources (e.g., literal and non-literal) as one could not know in advance all the questions or issues that would require a GEOINT response. Implicit in the definition is the desire to understand hidden patterns or correlations between related but unique sources. What unique intelligence can we gain from infrared, spectral, or SAR? Or, can we confirm with a greater degree of confidence a finding that we suspected?
Machine Learning: This technological advance is either a necessary post-condition of the explosion in GEOINT content or a pre-condition to ABI and big data analytics, or both. Increases in machine learning are used daily for all of the above reasons and more. For example, machine learning is used to increase confidence in an analytic result through triangulation of analytic results. Virtually any forecast for the explosive growth of GEOINT content in the next year or decade ends with the conclusion that organic assets (i.e., people) cannot keep up with the pace of information. Machine learning is the only practicable solution on the horizon.
NGA Director Robert Cardillo noted in his keynote address at GEOINT 2017: “For ‘Team GEOINT’—this is our time. We are standing where the SIGINT community stood when the internet became the digital fabric of the planet. And whether our new, persistent view of the world comes from space, air, sea, or ground—in five years, there may be a million times more than the amount of geospatial data that we have today. … We’ll either sink, or we’ll swim, or we’ll ride the rising tide. I say we ride!”
I could not agree more. The point is not that we won’t see new capabilities in the future—of course we will.
In keeping with the famous quote by Mahatma Gandhi, “The future depends on what you do today,” the GEOINT domain abounds with potential for new development.
There is enough yet unexplored territory in the current GEOINT landscape for us to make our mark, to reach new horizons, and to fulfill the promise of the GEOINT premise as envisioned not quite 12 years ago.