Benefits of Edge-Based Video Surveillance

Edge-based systems can also be less expensive, since there is no need for server maintenance, no proprietary software licenses to buy, and theoretically no barriers to scaling up the system, unlike server-based systems.

FREMONT, CA: There was analogue Closed-Circuit Television (CCTV) in the beginning, when computer analytics were not available, and it was up to those monitoring the images to spot as much as they could humanly do–a tall order when there were just one or two operators watching banks of camera feeds.  Then came IP video, which revolutionized surveillance by allowing video to be transmitted over a network cable and allowing IP cameras to be conveniently connected to networks.

Developers quickly realized that, now that video had essentially been digitized, digital video could be automatically analyzed using clever algorithms. This became a more common feature of increasingly sophisticated video management software (VMS), which could do things like detect movement or stationary objects, set virtual tripwires, and count people in retail environments, for example.

All of these analyses were previously performed at the heart of the video surveillance system, on the server or NVR loaded with the VMS. However, one significant disadvantage was the high cost of bandwidth required to transfer vast volumes of video to the VMS before they could be processed and analyzed.

However, in recent years, people have seen a move toward more analytics taking place at the camera end of the network, also known as the edge of the network. This method has the immediate benefit of allowing local analysis of camera images without having to send bandwidth-intensive video across the network. The analysis' findings are instead sent to the VMS as lightweight data. In practice, this means that a motion-detecting camera, for example, can only begin transmitting images if and when movement is detected. This has the benefit of reducing bandwidth and storage requirements, as well as eliminating the need for an operator to track video while nothing is happening.

The ability to record at the camera end is another advantage of this distributed network architecture. This allows for live video streaming at a low resolution to save bandwidth, while recordings at a higher resolution are made for post-event review and high-quality evidence. Of course, the recording system does not have to be in the camera; NVRs can be used to record for a group of local cameras at the edge. And if an edge system fails, it will only impact that portion of the network. Recording at the edge, with its distributed architecture, is especially useful in mobile and/or wireless applications, where bandwidth is much more limited.

Edge-based systems can also be less expensive, since there is no need for server maintenance, no proprietary software licenses to buy, and theoretically no barriers to scaling up the system, unlike server-based systems. The exponential growth of flash memory devices, such as SD Cards and USB drives, which can now carry many Terabytes of data, is making edge recording at the camera more feasible.