SD Still Kicking
Users are more likely than not to be interested in HD video. Just as a personal digital video recorder is to the average consumer, users certainly want their video surveillance cameras to produce more detailed images, said Xiang-Qun Ying, IP Camera Director for Hikvision Digital Technology. “However, the hefty investment in networking, storage and the cameras are prohibiting for cost-sensitive industries. For example, public agencies on a tight budget may have little choice but to settle for SD. Of course, there are also some scenarios where SD is already more than sufficient for the purpose intended.”
Smaller sites with relatively small fields of view and constant lighting tend not to fully need the benefits that an HD surveillance system can produce, Howie said. “Smaller sites with narrow confines, such as retail, banks and small offices, don't really have the distance/detail perspective to show significant value in using HD over SD.”
In some settings where only a small area is monitored — for example, a jet bridge — only a very limited scene needs to be covered, and D1 resolution would be sufficient to clearly recognize and identify a person, Ying agreed. “In other cases, you may need to cover a wide area, but not for petty details. For example, when monitoring for a forest fire, an alarm must be triggered when smoke is spotted. However, you don't need to see clear details of the flames. SD satisfies these applications in a more economic fashion compared to HD, in terms of bandwidth, storage and equipment.”
Other examples of prime SD applications are perimeter surveillance and locations involving looking in a large corridor, due mainly to the 4:3 aspect ratio providing better coverage of the scenes, Biemans added.
Looking at Big Picture
The popularity of HD video in the consumer market pushed forward video surveillance, but an exceptional HD video surveillance system requires much more than just good cameras. For IP-based HD video, things get a bit tricky. The edge device, transmission media and back-end processing all have a say in whether the system works properly.
Many of the system design considerations for IP-based HD video are similar to those for IP-based SD video, McCarrison said. “Cable distances must be considered, with restrictions easily overcome by good design and switch location, or the use of a fiber backbone. Many systems suffer from bandwidth issues when all video is streamed via a central server. This issue can be overcome by using a system with a distributed architecture, not only removing network bottlenecks, but also removing single points of failure and reducing costs.”
Edge Devices
Camera manufacturers could be a dime a dozen, but it is important to note that similar specs do not translate to similar performance; they will vary in image quality or bandwidth consumption. Good video encoding is critical to produce good quality video, and at usable bit rates. Networks that are already struggling to transmit HD video do not respond well to bandwidth spikes, which can be reduced through using a camera with better compression technology, McCarrison said. “As resolution increases, the ability to achieve good compression becomes much more difficult, especially when trying to maintain a guaranteed frame rate.”
In addition, the right cameras must be selected for the right locations. Each application and customer expectation of image quality should also be matched to the right megapixel camera, Zhou said. “While a 5-megapixel (MP) camera might be an excellent choice in some projects, it is not necessarily ideal for all scenarios. If an application requires full motion, for instance, a 1.3-, 2 or 3-MP camera would be a better option, or even a dual sensor day/ night camera. Being attentive to application requirements is a critical step in choosing the right megapixel camera and avoiding performance issues.”
Another area that can make a significant difference to the video quality is the lens choice. A quality megapixel camera requires a quality megapixel lens; without this combination, the benefits of higher resolution and good compression can be negated, McCarrison said. “Customers should not be tempted to save cost on the lens; otherwise, the money that they have spent purchasing an HD camera could be wasted. Some manufacturers spend a lot of time testing and selecting lenses, and often supply cameras with lenses that have been calibrated.”
An HD camera fitted with a standard lens will not deliver, as a dedicated HD lens capable of supporting megapixel images ensures top performance, Biemans agreed. “Indeed, the entire signal chain, from scene to screen, must be HD-optimized.” In the next part, we explore the chain of command in HD surveillance and common points of failure to watch out for.