Concern trolling? Maybe. But either way, the process of weeding through hours upon hours of satellite video for objects of “interest” just got a lot easier thanks to two researchers from Rochester Institute of Technology:
“It all comes down to efficiently handling large amounts of image data collected from satellites and video streams, which are not necessarily big images, but I can collect video for hours,” says Messinger, who also serves as the director of RIT’s Digital Imaging and Remote Sensing Laboratory. “You’d like to be able to download the data, have it go into a computer system and have it reduce that eight hours of video down to 20 minutes that somebody actually has to look at, just the highlights so they can process the information to make decisions.”
The RIT team’s work includes algorithms to analyze data in an “object based” way to better distinguish between noise and moving objects, determine their path, scale and other relevant data. The trick, of course, is that while humans live in a three-dimensional world, satellites can only capture two-dimensional video. Entire Hollywood special effects categories have been based on the fact that two-dimensional video can easily be manipulated to confuse the eye.
So, Rochester: let’s all relax about the red light cameras. Thanks to local boffins, you may soon be getting turn-by-turn navigation help from your friendly NSA agent.