Saturday, December 10, 2016

CCTV marries A.I.

Who will keep watch in isolated places?

I have been pondering security technology encroachments into public life, particularly regarding CCTV monitoring. There was a time, it seems now very long ago, that the UK began installing CCTV. Hundreds of millions of dollars, and over four million (and counting) CCTV cameras later, the UK is the most surveilled society on earth.

We were assured that would never happen in the US, or other developed countries. Violate our civil rights? No way cried the libertarian and democratic pundits in unison. 

Still, if you have nothing to hide…

Today, London and Beijing have over 400,000 CCTV each (proving politics is no guarantee either way). In the US there are over 30 million CCTV cameras, mostly in private hands. But there are now CCTV on streets in every major US city (Houston and Chicago lead the way with over 14,000 in Chicago alone) and public support is growing.

Will the brain behind the camera soon be an intelligent computer?


On one hand, we applaud when police apprehend the Boston terrorists due in large measure to public CCTV. We also later watched those same terrorists as they planted and exploded the devices - prevention was not a result of those cameras. 

I always applaud traffic intersection CCTV to cut car crashes, especially in my city where drivers spend more time in narcissistic self-obsession trying to beat the red light rather than watching where they are going. 

CCTV in public spaces is not new


Recently I’ve been reviewing the latest in CCTV analytics, intelligent tracking and real-time scene analysis. In other words, CCTV on steroids. The latest is no longer motion detection or auto tracking (so old school). The latest is intelligent video analytics, a major evolution from facial recognition software of yesteryear. Video analytics is made possible by exponential increases in processing power and so-called ‘intelligent’ algorithms. 

And now it is part of security and public safety, watching for suspicious movements, packages, behaviors. Watching you! How does the computer know what to look for? It uses algorithms based on past behavior. In future, it may use artificial intelligence to learn on its own. And that is where things get interesting.

Wired magazine puts it this way:
“voice, image, and motion recognition will transform human-computer interfaces into a seamless interaction between the user and all the computing devices in that person’s life.”
A few years ago I blogged about economic theorist Jeremy Rifkin, his forecasts, the Internet of Things and his predictions for disruptive technologies. It seems he was right. Should we be worried?

- Gregory Saville