provided by: 
Video analytics, also referred to as machine or robot vision, is embedded in a component such as a DVR or NVR, software video engines in a surveillance system, or video management and networking equipment typically deployed in infrastructure security, personal safety, and retail sales and research.
Video analytics is software designed to evaluate real-world situations, known as behaviors, typically the analysis of events such as unattended objects, unattended vehicles, interactions between people and tailgating. Even a single camera or channel will have a number of behaviors being processed simultaneously to meet a given requirement.
The deployment and acceptance of video analytics is limited by platform cost performance and the impression held by many that it is still in a beta stage. But in my opinion, innovation is only achieved by pushing the limits.
As is the case with other types of security, maximizing probability of detection also raises the probability of false alarms.
Generally speaking, video analytics operate most successfully on less complex analysis, including whether or not an object or a human crosses a perimeter, people counting, reverse directional flow (walking the wrong direction in a queue) and tracking objects.
Many video products offer a degree of video analytics such as motion detection and direction sensing and search criteria. Many of these functions are here today and deployable. However, exaggerated marketing claims and wishful thinking may combine to create an environment of disillusionment and a bad rap among some industry insiders.
Is this another case of more bleeding edge technology? I say, "no pain, no gain."
As a relatively new technology, there is an absence of standards in measuring the actual performance of video analytics applications which represents another hurdle to its widespread industry acceptance.
But of course, there are reasons as to why there is so much interest in video analytics. I've examined some of these reasons below:
Increased productivity without increasing staff. A system employing analytics can monitor hundreds of cameras simultaneously and supply only alerts and relevant video to a security guard for specific situations requiring their attention.
Forensics in applications. To enhance the use of video files in legal proceedings, video analytics can provide added information in the form of date and time stamps.
Technology improvements are here. Real time as well as linear processing of one task is possible due to the steady improvement of processing power. New network processors and DSPs have opened new opportunities for growth in the industry.
Distributed intelligence across networks. Intelligent video routers, which will become the next major security platform, interface with cameras and route live video and alerts to central servers and storage devices, and collaborate with access control systems providing a real-time interface between a physical security network and the IT network.
SECURITY DEALER TECHNICAL EDITOR TIM O'LEARY is a 35-year veteran of the security industry and a 15-year contributor to the magazine. O'Leary's background encompasses having been a security consultant since 1986 and an independent security company owner/operator, in addition to his research and evaluation of new technologies and products introduced to the physical and electronic security fields. He is a member of the VBFAA (Virginia Burglar and Fire Alarm Association); certified for Electronic Security Technician and Sales by the VADCJS (Virginia Department of Criminal Justice Services); and, has served as a judge for the SIA New Product Showcase. Send your integration questions to Tim.Oleary@secdealer.com.
author: By Tim O'Leary