Analyzing crime
Because of a growing demand for security technologies to improve public safety, many city and county governments have installed closed circuit television (CCTV) surveillance systems to monitor public facilities, high crime areas and potential terrorist targets. Many of the systems consist of hundreds, even thousands, of cameras. Despite new technologies and system architectures that manage the ever-increasing volume of video, public-sector security has not improved significantly.
Although some criminals who see cameras may be deterred, security departments also must be able to see threats to respond quickly. Traditionally, that means a guard must watch security camera monitors, but that practice has several limitations, including the extremely low probability that a guard actually will see suspicious activity or a crime in progress. Studies show that even trained security professionals get bored, tired and distracted while watching security video. Also, as the number of cameras in the system increases, more security personnel are needed, increasing costs. And, if security personnel are continuously monitoring CCTV, they cannot be responsible for other tasks.
To support traditional monitoring efforts, cities and counties are using video analytics software to store and retrieve critical video for real-time incident recognition, assessment and response, and for forensic purposes. More than capturing motion, the software continuously analyzes images from security cameras to identify unusual events or situations, such as the presence of people or vehicles within secure areas, objects left behind or a mugger lurking in the hopes of catching a lone passerby. Motion detection devices are similar to sensor technologies; they can detect motion but cannot provide additional information. Video analytics can describe the type of event and connect it to video of the incident. For security operations with limited resources, enhancing a video surveillance system with video analytics allows small security forces to proactively and cost-efficiently monitor large video surveillance installations.
When video analytics software detects an event, it sounds an audible and/or visual alarm and can send text message alerts or video clips through e-mail or cellular phones to security personnel or police officers, allowing them to respond quickly and possibly preempt an incident. The alerts and files include details about camera location, when the event occurred and the detected behavior.
For forensic purposes, video analytics help law enforcement officials sort through thousands of hours of security camera footage quickly. When an alert is triggered, the video clip is tagged with information concerning what happened, when and where, so officials can find the right video segment to use in their investigation or prosecution.
Analytics can work with almost any CCTV system, including analog and IP, and many systems do not require additional hardware. Analytics-enabled video surveillance systems make crime detection, risk management and enhanced operational effectiveness a reality. Video surveillance installations are substantial investments, and analytics-enabled systems help civic leaders maximize public safety for residents as well as the overall return on the investment.
The author is president and CEO for Reston, Va.-based Cernium Corp.