ISCAPS Aims for Intelligent Surveillance

Share it on Twitter  
Share it on Facebook  
Share it on Linked in  

Usually, a blog named Gadget Envy deals with equipment that is useful for business or is just plain fun. In this case, the issue at hand is more serious -- even a case of life and death.

The descriptively named Integrated Surveillance of Crowded Areas for Public Security (ISCAPS) project described in this CNET story has just concluded. The $311 million, 10-company European initiative researched and developed intelligence security cameras that proactively identify potential problems. Clues the system will flag range from loiterers to abandoned packages.

Spotting something that could lead to potential trouble is a difficult thing to do, especially when it is done in an airport, a train terminal or some other busy venue. The thinking is that the use of sophisticated cameras -- or, more accurately, the software programs behind them -- will increase the odds of preventing a disaster. The trick is developing software that has the smarts to fish suspicious activity out of the billions of images recorded by the cameras.

The research focused on several activities, such as erratic movements and loitering, that could be signs of trouble. ISCAPS's work included interviews with experts and role playing with actors. ISCAPS also created statistical models defining the parameters of normal behavior during various times of the day and what constitutes a deviation from those norms.

Of course, the final decision on whether the person running down a train platform is a terrorist or a tardy commuter is made by the human monitoring the system. This new generation of cameras, however, clearly will make that job -- which must be done in a split second -- easier. ISCAPS also worked on technology that will help track people after they do something that raised the software's electronic eyebrow.

ISCAPS and similar projects are deeply related to the new world of Internet-based digital surveillance. Suppose that a crime is committed and the only lead is that a witness remembers seeing a man in a blue cap at about the time the event occurred. Formerly, it would take a tremendous amount of time and effort to sift through surveillance tapes looking for a blue-capped man. In a digitized environment, the right software can perform this task in seconds. The market for IP-based surveillance appears to be growing.

The use of IP networks also is significant. It revolutionizes things in much the same way that it changes corporate communications. In this environment, public safety officials rushing to the scene of an emergency can have any information -- digitized photographs, building schematics, directions and alternate routes, etc. -- sent to them in transit. Also, vital information -- a notice about the blue-capped man, for instance -- can be distributed worldwide in seconds.