We wrapped up an amazing experience in San Antonio last week at USGIF’s GEOINT 2017 Symposium. This flagship event for the geospatial industry brings together our community to reflect on where it has been and look forward to where it is going.
In past years, it felt as though the GEOINT community was focused on evangelizing the potential of disruptive technologies like cloud computing, pervasive sensors and big data analytics. This year, it was evident our customers are seeing early mission wins enabled by innovation and recognize the potential to transform the intelligence community by riding the wave of commercial technology, which is accelerating at an exponential pace.
U.S. Defense and Intelligence Leadership Recognize the Need for Change
During the GEOINT keynote presentations, NGA Director Robert Cardillo, DIA Director Vincent Stewart, and Director of Defense Intelligence for Warfighter Support Lt. General Jack Shanahan spoke about the opportunities and threats associated with the proliferation of sensors and the resulting explosion of new data sources. With thousands of new satellites on orbit, millions of UAVs and billions of mobile devices, we will collect enormous amounts of information about our changing planet. Interpreting all this data manually would be incredibly expensive and too slow to provide a source of strategic warning. The convergence of these factors has created a sense of urgency across the defense and intelligence community to create a force multiplier for scarce analyst resources through automation, crowdsourcing and big data analytics.
Lt. Gen. Vincent R. Stewart, Director of Defense Intelligence Agency
They See Promising Results from Disruptive Technologies
In his keynote speech, Director Cardillo referenced two projects—Beachfront and NOME—that our developers and analysts at DigitalGlobe | Radiant are honored to support. Beachfront is a means of generating automated shoreline data using a variety of imagery sources. The NSG Open Mapping Enclave (NOME) is a crowdsourced mapping platform that enables NGA customers on multiple security domains to extract roads, buildings and points of interest from satellite imagery and share data in the cloud. In these two clips, Director Cardillo shares how Beachfront and NOME are helping NGA extract information many orders of magnitude faster and cheaper than conventional methods.
Robert Cardillo, Director of National Geospatial-Intelligence Agency
At Point of Arrival, These Capabilities Can Transform the Mission
The government is recognizing its tremendous opportunity to take advantage of commercial innovation. Director Cardillo stated in his keynote that he aims to automate 75 percent of the repetitive tasks analysts perform in order to allow analysts to focus their time on higher order synthesis and pattern discovery. For this vision to become reality, our customers need to train effective algorithms to automatically recognize objects and activity; create scalable methods to enrich, store and index all of this data spatially and temporally; and develop analytic modeling tools able to reveal patterns by processing massive amounts of data. Our investments to advance research in the application of machine learning to remote sensing data to enable automation through open innovation initiatives like SpaceNet and our free Software Development Kit (SDK) DeepCore will play a role in realizing this future.
DigitalGlobe Has Developed Robust Capabilities at the Heart of These Trends
DigitalGlobe has been a pioneer in harnessing technologies like cloud computing, crowdsourcing, machine learning and big data analytics to transform global mapping and intelligence. We are delivering innovative solutions built to scale in the cloud that combine unique IP, an open architecture and an ecosystem-friendly business model.
Over the next few weeks, we will publish a series of blog posts to share how DigitalGlobe is leading initiatives that respond to the calls to action outlined by leadership at GEOINT and will move our community forward:
- Automation is the way of the future, and there are many commercial and open source capabilities that can be applied to GEOINT data to help intelligence analysts scale.
- Machine Learning requires an enormous amount of training data to yield effective results. We are leveraging our vast imagery archive and several patented technologies to create massive training data sets we can make available to accelerate algorithm development.
- DigitalGlobe has leveraged crowdsourcing since 2010 as a way to rapidly analyze satellite imagery. We are now leveraging this capability to rapidly validate machine learning algorithms to make them smarter over time.
- In our constantly changing world, current maps are key to mission success. DigitalGlobe is uniquely positioned to harness the explosion of public data sources to create foundation data sets like Human Landscape and collaborative mapping applications that create a living inventory of features on the planet.
- By using these various methods to create the DigitalGlobe, we now have the ability to enable analytic modeling for anticipatory intelligence to help analysts narrow the search space and gain early warning
This was DigitalGlobe’s first GEOINT Symposium since our acquisition of Radiant. It was terrific to see team members who in prior years represented DigitalGlobe, RadiantBlue and HumanGeo share the vision of our combined services business, DigitalGlobe | Radiant. Working together with our Platform and Imagery business units, we want our customers and partners at GEOINT to realize DigitalGlobe is a mission partner that harnesses disruptive technology to deliver innovative solutions that sit at the intersection of national security and GEOINT tradecraft.