Sunday, February 03, 2008

Hitachi CCTV system will identify troublemakers by itself

Hitach-CCTV-system
When it’s not mandatory you may find pleasure (for the ones who love kinky stuff that is) in watching videos streamed by surveillance cameras in public places like malls, restaurants etc. Now imagine, doing the watching on everyday basis, hour after hour, day after day. I salute those security personnel (both Federal and private) who do this enormously tedious work diligently 24*7, for us, for our safety. But they are human too, and human are vulnerable to errors and unmanageable data, remember they are bombarded with steady inflow of videos, and God knows in a giant mall or hospital on how many number of TV sets they have to fix their attention on. As companies worldwide are spending more and more on security accounts, it is becoming logistically impossible to make sense of the overloaded CCTV streamed information.

Now, realizing the vacuum in the security sector, Hitachi has come up with an innovative solution. This new system by using artificial intelligence will evaluate what images these CCTV cameras are capturing; it will then sift and prioritize selected data for the human inspection.

The system takes feeds from up to 100 online cameras and passes them to a personal computer for image processing and a central server for the meatier task of searching through video archives for comparable footage.

Together, they pull out the four most important video streams and display those on a monitor at high resolution, while the remainders appear as small thumbnails. Source

All that is fine, the system will use face recognition software to find out if the repeat trouble makers are in to create some more trouble. But what about suspicious criminals or terrorists (who may not be recognized by the system as criminals), can the system be able to identify them as well, will the system be able to detect first time suspicious souls who may look vulnerable to an experienced naked eye but not to a software?

No comments:

Explore more

Custom Search