The IT Perspective On Video Analytics

Mar 1, 2008 12:00 PM, By Nik Gagvani

As IT professionals weigh in on video analytics systems, they should consider bandwidth issues and network configuration.

It is now common knowledge in the security field that legacy surveillance systems can be greatly enhanced by applying video analytics and intelligent video technology. The ability to automatically detect events of interest and to alert security personnel to these problems in real-time has greatly enabled security departments around the world to better police their facilities. The benefits of video analytics, however, are not just limited to the security department.

video technology. The ability to automatically detect events of interest and to alert security personnel to these problems in real-time has greatly enabled security departments around the world to better police their facilities. The benefits of video analytics, however, are not just limited to the security department.

While the responsibility of selecting, purchasing and monitoring a surveillance and/or video analytics system lies with the security director, the actual installation and day-to-day upkeep of the technology falls to the IT department. Network administrators can be put in somewhat of a bind when faced with the resulting video traffic from these devices, especially as it relates to bandwidth provisioning and management and data storage.

Video traffic is not a new concept to network administrators and IT departments. Almost every network is set up to handle some kind of video load, usually Web-conferencing or a limited set of video content, which is accessed by users across the network. Surveillance camera feeds, however, do not fit into the realm of “common” video traffic. The cameras are running and recording continuously, but the information sent may not be viewed immediately or at all, and then only by a small group of users (usually, the security department). This creates a massive confluence of video traffic that, when all is said and done, may not actually be that important.

Another issue stemming from video surveillance is storage. Video files take up far more room than standard text or image files, which, when coupled with the round-the-clock nature of the cameras, can make storing video extremely expensive. Network administrators need effective ways to manage the bandwidth consumed by surveillance feeds to store important video content.

Video analytics and intelligent video systems can solve these surveillance-related problems for IT departments. Essentially acting as a video “filter,” intelligent video systems ensure that the right video reaches the right people at the right time, and that only important relevant video content — such as segments needed for future forensic review — are stored.

From a bandwidth standpoint, analytics drastically reduces the traffic from surveillance feeds. Since a camera or encoder with video analytics monitors continuously for certain pre-determined behaviors, such as lurking individuals or left packages, video is only transmitted when these behaviors are detected. Video analytics can do a far better job than motion detection does because it “qualifies” the motion prior to transmission and cannot be “confused” by moving foliage and other environmental factors. The resulting freed-up bandwidth can be used to stream relevant video at a higher resolution, higher frame rate and/or better quality.

This same filtering mechanism of video analytics can also help dramatically reduce the storage problem. With only relevant video being stored, there is a dramatic reduction in storage requirements. There is also a multiplier effect, resulting in lower power costs, space and personnel, to be derived by implementing intelligent video into an existing surveillance system. Further, an added advantage for consumers of surveillance video is reduced search-time. Incidents can be located instantly, and security personnel can spend more time solving and preventing crime rather than in reviewing video.

With these benefits in mind, when weighing in on the security director's decision-making process on selecting an analytics solution, what should IT departments consider? There are three key considerations within an analytics system that network administrators should be aware of to provide the best solution that meets security requirements.

1. Location — Analytics can be employed at the edge of the network, in cameras, encoders and processors, or in a centralized location. Positioning analytics processing close to the edge offers the greatest saving in bandwidth; however this needs to be weighed against the maintenance and setup costs for multiple edge devices. Each facility has a unique network topology and needs. Frequently, analytics processors can be placed in a local network closet, which is sufficient to throttle bandwidth usage over long-haul links. A standalone analytics processor, or a smart encoder, allows security personnel to leverage existing cameras and video infrastructure instead of replacing it with new intelligent cameras.

2. Setup — Network administrators should look for an intelligent video system that integrates seamlessly with the current network and allows for some form of remote administration and management. The last point is especially valid in distributed situations, where one IT department may oversee multiple sites with analytics installed. Configuration changes must be possible from a centralized location, and this configuration must be capable of being pushed automatically to multiple processors or cameras.

3. Knowledge Base — IT departments need to select a system that requires little “security knowledge” and that, from a network standpoint, behaves similar to any other device that is attached to the network. This includes setup, configuration, updates, health-monitoring and port configuration.

With the rapid transition to IP video, the network will remain a focal point of security for the foreseeable future, adding to the responsibilities of the IT department. By using video analytics to their benefit, network administrators can not only keep the network up, but play a vital role in keeping the enterprise safe as well.

Nik Gagvani, Ph.D., is vice president of engineering and CTO of Cernium Corp., Reston, Va. Cernium develops and markets video analytics-enabled products for security, surveillance and business intelligence.

Want to use this article? Click here for options!
© 2008 Penton Media Inc.

This month in Access Control

• Opening Up About Door Closers
• An Enterprise Approach
• The Framework For Open Systems
• On A Higher Plane
• More from April's issue