Can your system support that new 4k camera?
The security industry is easily whipped into a frenzy by the latest shiny new gadget, and advances in 4k camera technology are generating a loud buzz. However, I’ve been discovering why it might be best to take a step back and assess whether you really need to be one of the first to boast a 4k camera or two in your surveillance solution.
What is 4K?
4k is a solution that offers four times the resolution of 1080p HD, which in the movie business means crystal clear clarity and a life-like viewing experience. 4k video runs at 4096 x 2160 pixels, and can offer a range of benefits for surveillance, like the ability to reduce the number of cameras you need, and the ability to zoom in for close detail without the image pixilating beyond recognition. These are just a couple of the reasons camera manufacturers a lining up to release 4k products, and all indicators are that end-users are excited by the new tech.
However, the more I dug into the subject, the more I discovered that there’s more to it than simply plugging in a 4k camera and enjoying the benefits – you’re going to need to upgrade more than your cameras.
Storage, network and transmission
All those extra pixels mean bigger, heavier images to move around and store, so before you start comparing cameras you need to look at what your system can handle.
Under new compression standards for 4k encoding Ultra HD resolution requires at least 12 Mbps bandwidth. By comparison, the standard 1080p HD currently being widely used only requires 4 Mbps. 4k puts pressure on the network as recording in the highest resolution (and why would you spend all that money on a new 4k camera and NOT record to the highest resolution?) means hard work for your system to stream from your camera to your network video recorder (NVR).
Streaming 4k over standard ADSL is simply not effective at full frame rate. However, there have been technology advancements in terms of Ethernet, seeing the development of 40 gigabit Ethernet becoming available. This enables the transfer of Ethernet frames at speeds of about 40 gigabits per second, which is suitable for 4k.
Essentially, are your pipes big enough? If not, add an upgrade for your network to the shopping list.
And while you’re looking at that, take a look at your graphics card. Can it handle the significant increase in the flow of data? This is especially important when assessing the technical challenges posed by streaming from multiple 4k cameras.
Once you’ve got the capacity to move the images around, you’re going to need somewhere to store them. Four times the resolution means four times the storage. If you’re planning on investing in a camera that includes H.264 compression within the device, such as the 4k offering in the pipeline from Arecont Vision, storage may be slightly more straightforward.
Scott Schafer, Executive Vice President Sales and Marketing at Arecont Vision, outlined the company’s plans for a new 4k product, and he insists that although storage needs will increase, H.246 does make things more manageable.
“We’re excited about going into production with our new 4k product this year and it’s going to be a big hit,” he said, “It’s often said that storage is an issue for 4k technology, and as an 8 megapixel camera it will take more storage, but with H.264 compression we’ve got this running in a manageable way. So we have a practical solution – we know it won’t work for every solution, but it will be good for a lot of environments.”
Now you can move your video through your network and you can store it. But what about getting that Ultra HD footage to third parties? One of the top selling points of 4k is how beneficial the amazing clarity of the footage will be for evidentiary purposes. You may have invested and boosted your own network, but has the local police station or courthouse? If the footage is to be used in court, have they wrestled out funding for a suitable 4k monitor to view it on? All points to keep in mind if your main motivations for considering 4k are forensics related.
Lenses, low-light and the control room
Let’s talk about lenses. A true 4k camera requires an 8-megapixel lens, and these are not cheap – nor are they readily available with only a couple of manufacturers currently producing lenses to this specification. In addition, they are considerably larger than their lower megapixel cousins and this has a direct affect on the size and style of camera. As 4k cameras need to be physically bigger to accommodate these lenses, it means that it’s not currently possible to get your hands on a 4k compact dome camera – a favourite in the surveillance industry.
Also popular in the sector are cameras that work well in low-light environments. This is an area where existing 4k technology falls down. The result of small sensors in 4k cameras and the volume of pixels in a frame is less light being absorbed. Julian Rutland, Director – Visual Communication Products & Solutions at Canon Europe Ltd., explained that this is one of the reasons Canon is holding back from 4k for the time being.
“When we talk to customers about our existing and future products there are questions about resolution,” he said. “However, if you get into a deeper discussion and ask what they really want from their camera then additional issues like low-light performance come up. When you go to a high-resolution sensor pixel size drops and low-light sensitivity also drops – and that’s one of the things that customers are really looking for. So you have got a natural trade off between resolution and low light performance, it doesn’t matter how you look at it it’s there.”
So far we’ve looked at the cameras and how you move and store your Ultra HD footage, so that just leaves viewing the footage, which brings us to the control room.
Strictly speaking, you don’t need a 4k compatible monitor to enjoy the benefits of zooming into the face of someone trespassing on your premises, but you’re not really getting your money’s worth if that’s all you get from your 4k device are you? To really ‘see’ 4k you need the proper monitors, and while LCD 4k monitors are becoming more commonplace in the commercial market, the fact is they are not yet suited for use in surveillance.
Even then, the typical user interface in a control room centres on viewing more than more camera on one monitor, only jumping to a full screen view from one camera occasionally. In multi-view you’re not going to see any difference in footage when compared to a 1080p HD camera.
Early days
The crux of the matter is that 4k technology is very new, so there’s a lot that needs to advance and develop before it’s truly beneficial to the surveillance market. There’s no doubt that one day 4k will become the standard for the industry. But at the moment, no one is in a position to offer a full 4k solution, and the advice seems to be to think long and hard about whether you really need 4k right now – especially with solutions available that will work well with your existing systems.
“There are a small number of applications that need a higher resolution but they’re quite specialist at this time,” said Canon’s Rutland. “As a company we will embrace high resolution, because high resolution is a driver of quality, but there is more that we can do with full HD resolution, we don’t need to push the sensor and have all the bandwidth issues that people are talking about. With our lens technology and our image processing we are getting market leading image quality without the overhead of bandwidth and the other concerns that come with going to 4k.”
Dieter Dallmeier, Founder & CEO of Dallmeier, added, “The technological advances that have been made in the field of network-based surveillance in the last few years are astounding. But resolution on its own is not everything. Many megapixels on a single chip have unpleasant side-effects as far as things like light sensitivity, lens quality, image rate and bandwidth are concerned. What really interests the customer is not the number of megapixels, but rather whether suspects can be singled out and identified – regardless of which area of the surveillance field that person is located in. So it is not the resolution of the camera itself, but how many pixels per metre are obtained for the relevant object or person. This approach is precisely what we are following with our multifocal sensor technology – a completely new and already patented camera technology.”
