The Disturbing World Of Closed Circuit Television and Facial Recognition

Closed Circuit Television or CCTV has proved its value as a way to combat crime. You would think that half a million cameras would be an adequate line of defense for a city like London. However there are limits to the effectiveness of two dimensional CCTV images and facial recognition software. The challenge for future surveillance now is not to just to identify a face in the crowd but anywhere in the world.

London is the most heavily surveyed city in the world. More than half a million cameras keep watch over the Great Britain capital. The center of London is surrounded by a ring of cameras. Within seconds every plate number is fed into a computer that tracks the comings and goings of each motor vehicle. You would think that the many eyes of Big Brother would be an adequate line of defense as a surveillance weapon.

Closed Circuit Television or CCTV has already proved its value as a way to combat crime. In August 2004, a group of men attached an electronic device to a bank machine in London. They somehow managed to steal card and pin numbers and withdrew cash from those accounts over and over again. CCTV was used to capture these men. The men behind CCTV followed these men and the police were able to apprehend the culprits before they managed to escape.

There are two critical limits to closed circuit surveillance. You have to know who you are looking for and you have to have a human watching every screen every minute of every day. The challenge for future surveillance is finding people before they commit crimes. A task no simple camera can achieve.

Every year more than 35 million visitors visit the city of Las Vegas Nevada. Many people know that the Las Vegas casinos use cameras but few realize the extent to which they are being watched. Gamblers spending six billion dollars here every year and that is why casinos use the most highly technical systems in the world to monitor their clients. Their aim is to catch someone doing something illegal or out of the ordinary that is going to protect the property and Casinos take that very seriously.

The cameras may look simple but from forty foot ceilings but they can track a card player who continually wins at the tables. If the player has been banned before then the player would have been previously caught on tape. To find out they use facial recognition software. They take time lapse snapshots of the players face on the monitor and feeds that information into the software program.

There are eighty primary measuring points on the human face and if the photograph is going to work they need the right angle and lighting. The computer describes the face but plotting these points and measuring the variants between the features. These points create a unique numerical code called a face print. The computer compares the subjects face print to a database of known undesirable gamblers from casinos around the world. Despite this technology, facial recognition systems still rely on high quality images and the final decision relies on human decision making.

Facial recognition experts are aware of the limitations of the standard two dimensional facial recognition system. While it can see through basic disguises it cannot identify people from the side. If the subject simply turns his head the computer does not recognize him because it can no longer measure the distance between the features.

On Wednesday morning 7th July 2005, three million commuters caught London cabs, buses and trains. Closed circuit television operators kept an eye open for any suspicious behavior. At 0850 hours, a series of explosions tore through three subway trains within a minute of each other and less than an hour later the top floor of a bus exploded. Fifty two people died and seven hundred were seriously injured that day in one of the worst acts of terror in British history. In this case of the London Bombings, even if the suicide bombers photos had been in a database, investigators would most likely not have made a match. The images were too blurry and shot from strange angles.

Facial recognition experts know that a more advanced system could help in the future. Scientists are developing a new surveillance system that take facial recognition into another dimension connecting three dimensional imagery to satellite tracking. The challenge for future surveillance is not to just to identify a face in the crowd but anywhere in the world.