While traditional satellite imaging companies like DigitalGlobe continue to corner the market for best quality imagery, it’s the rise of small satellites and the mass quantities of data they produce that is the future of the industry.
Simply put, small satellites just don’t have the apertures to produce high-resolution imagery. But what they do have is the ability to get more frequent imagery— think data sets that cover the entire Earth daily. As it becomes less and less expensive to build and launch small satellites, investors are capitalizing on those reduced costs and promise of big returns on the data these satellites will produce.
“We’re not competing for perfect imagery,” said Carissa Christensen, Managing Partner of the Tauri Group. “We’re competing for imagery data that can be run through an analytic platform and interpreted for business insight.”
Every year, satellite industry leaders and global development experts convene in Washington, D.C. for the annual SatSummit to discuss how satellite data can be used to solve some of the world’s most crucial challenges — from climate change, to population growth, to natural resource availability.
Major satellite data providers including NASA, ESA, DigitalGlobe and Planet were on hand, along with big data gurus from AWS, IBM, and Orbital Insights to discuss some of the unique challenges associated with satellite data. One such challenge is the concept of “fat data.” Imagery files are bulky, and create problems for those tapped to archive, run analysis on, or distribute that imagery data on the web.
As more satellites are launched — 85 in 2016 alone, according to Melanie Preisser of Vulcan Aerospace Corporation — more and more data will continue to pour in, and there simply are not enough imagery analysts in the world to analyze it all. Advances in machine learning algorithms and other big data technologies are helping to extract insight from the data for use in a variety of applications. As demand for data continues to grow, the focus has shifted from the satellites themselves to software and services that can be derived from the data.
OGSystems’ Blue Glass; platform is at the forefront of the budding analytics-as-a-service industry. By partnering with best of breed companies in commercial satellite imagery, natural language processing, machine learning, and predictive analytics, OGS created a true space to ground intelligence platform that will deliver the business insight industry craves.
“What I believe the value for someone like OGSystems or another analytics provider is going to be is identifying how we understand our customers enough to logically array disparate content to give them a solution to solve their problems,” said John Goolgasian, OGS Associate Partner leading Geospatial Analytics.
Through its development of Blue Glass, OGS is building the opportunity to apply machine learning and artificial intelligence to structured and unstructured content to determine normalcy, alert anomalies, and generate automated intelligence reports.
“Blue Glass is doing the hard data work that today takes analysts countless hours to perform, and we’re automating data analytics and narrative generation to put in front of analysts and operators who can use it for further investigation and for operations,” Goolgasian said. “We have already begun to generate basic analytic reporting and will soon implement a compelling approach to trend and pattern recognition with the goal of predicting likelihood of activities occurring across space and time. With automation, I still believe you need to present this information to someone to make an intent decision. I don’t think that machine learning can do intent yet. We want humans to be in the loop, training the models and deciding the next course of action.”