Persistent surveillance relies on extracting relevant data points and connecting the dots
THE MIL & AERO VIDEO BLOG, 28 Aug. 2012. Military persistent surveillance can distill actionable intelligence from a virtual ocean of data. In an era of shrinking military budgets, this technology is still considered part of the most promising technological areas in an otherwise dwindling U.S. Defense Department spending plan.
THE MIL & AERO VIDEO BLOG, 28 Aug. 2012.Military persistent surveillance can distill actionable intelligence from a virtual ocean of data. In an era of shrinking military budgets, this technology is still considered part of the most promising technological areas in an otherwise dwindling U.S. Defense Department spending plan.
Military persistent surveillance has been receiving a lot of attention lately, and I can see why. Persistent surveillance can distill actionable intelligence from a virtual ocean of data, and rarely puts human lives at risk, because more often than not, persistent-surveillance platforms are unmanned.
In an era of shrinking military budgets, command, control, communications, computers, intelligence, surveillance, and reconnaissance -- better-known as C4ISR -- is still considered one of the most promising technological areas in an otherwise dwindling U.S. Defense Department spending plan.
We hear a lot about persistent surveillance these days, but until recently I never really pondered what the term means. At first glance, it sounds like a police stakeout -- guys with binoculars concealed in parked cars or in motel rooms stifling yawns amid a clutter of empty paper coffee cups, crumb-filled doughnut boxes, and crumpled cheeseburger wrappers.
We've all seen it in the movies: police, after hours of boredom, finally see the drug deal go down. A quick radio call orders other police to move in, the bad guys get busted, and everyone goes home happy.
But it turns out there's more to persistent surveillance than that -- much more. It's really about extracting relevant data points from hours, sometimes even weeks, worth of imaging data, and then connecting those dots to create a viable intelligence picture.
Here, let me explain.
Say intelligence analysts somewhere in Afghanistan are studying persistent-surveillance imagery taken from multispectral sensors mounted to a tethered aerostat, an unmanned aerial vehicle, or even a manned aircraft.
First, they take note of motion, be it cars, groups, or even just a person walking alone. Motion, of course, is a tipoff that something of interest might be happening. Motion is one data point, and a reason to focus in on portions of the imaging data containing motion.
Focusing on that motion might reveal concentrations of cars parking at a particular place all within minutes of one another, where cars hadn't been converging before. Might it be a terrorist meeting, or the planting of a roadside bomb?
Maybe. But the wealth of persistent surveillance information can enable analysts to concentrate not just on what's happening at that moment in the video data stream, but also to rewind and see where the suspicious cars came from, and move forward in the data stream to see where those cars went after leaving the area.
That creates more data points, and more dots to string together into what may become a coherent picture. Did the cars come from or go to a suspected terrorist hideout? If an IED subsequently explodes in the vicinity of that car gathering, this ability to rewind and fast-forward recorded events might help locate others involved in some nefarious plot.
It's almost as if persistent-surveillance technology is leading to the ability to compress time. Now it's up to industry to come up with the right kinds of software tools to help intelligence analysts find those dots and connect them quickly.