University of Bristol
Institute of Physics logo
Why not try our other site: BEEP Biology & Ethics

The future of CCTV

Automated people tracking or only watching the baddies?

Vast hours of CCTV video are recorded every day. In the past, when CCTV has helped to solve crimes, it is has been because a human has studied thousands of hours of recordings to spot a suspect.

But now, facial recognition software can automatically analyse video, pick a face from a crowd, compare this to a database of faces and discover who the person is. This can dramatically speed up the task of sifting through huge quantities of video.

The latest research into automated surveillance goes even further. It looks not only at faces, but analyses behaviour such as how you walk. Computers can follow the same person from camera to camera across town, without any human involvement at all. Given the sheer numbers of cameras in town centres, this is only a short step away from being able to automatically track a person wherever they go.

What is the difference between a camera following your every move and a person following behind you everywhere you go?

Question:  Do you think this is acceptable in a ‘free’ society?

There are calls to make surveillance cameras only record potentially 'bad' behaviour, rather than everything they see. Find out more

Activity: Statistics and rumours:

How many cameras are there really and how can we know?
Read this article that challenges the numbers.

There are many types of cameras, in private companies, used by the police, or just in a corner shop. How realistic is it to fear that they could all be linked up to follow individuals?