HDs Defining Moment
Posted by Security Electronics and Networks | @Articles CCTV | June 16, 2011, 7:00am AEST
FOR many years establishing nominal CCTV camera performance was a fairly easy task for most end users in the era of analogue. You simply bought the camera with the highest horizontal resolution and the lowest minimum scene illumination.
It goes without saying that these numbers were no guarantee of real performance. It’s easier to print exaggerated claims on a brochure than it is to build a 640—line CCD (or winkle one out of Sony). As IP cameras appeared on the market the complexity of establishing real performance has become more difficult.
Signal paths now include digital processing, compression and network components – all of which pose challenges. Then there’s the trend towards CMOS sensors in place of CCDs. While the performance of CMOS is incomparable to the sepia-toned clunkers being sold ten years ago, CCD is still optically better in low light and has less digital ‘spin’ applied to it.
With IP, megapixel and HD cameras on the market instead of becoming easier to assess performance, it’s become harder, with multiple variables now muddying the waters to the detriment of many dedicated high end CCTV manufacturers.
No one would argue a fixed standard would be beneficial by where might such a standard come from? Industry associations are not in a position to handle this requirement – it’s too complex and expensive. The way forward has long been an industry standard inherited from consumer electronics.
So – with the introduction of the consumer HD TV standard (720p and 1080p) to the surveillance industry and the release of SMPTE’s HD standard we’re certainly entitled to wonder whether or not finally there’s a standard users and installers can depend rely on to accurately attest to performance.
Perhaps it’s a little early in our tale to make a judgment on HD as the default standard but my personal opinion is that given the open-ended nature of SMPTE’s standard and the multiple possible interpretations, we are closer to an objective CCTV standard but not close enough for end user comfort.
As part of research for this story, I spoke to leading manufacturers and distributors in order to get a feel for where they feel the industry is where HD is concerned. The general feel is positive but there’s also recognition that HD is an interpretative standard.
HD is variable
First up is Avigilon which manufactures a full range of HD cameras, from 1MP to 16MP, variously using H.264 and JPEG2000 compression. The company also has an encoder allowing its ONVIF-compliant NVR management solution to adopt legacy cameras. The high performance Avigilon range includes a 180-degree and 360-degree 8MP camera.
One of the characteristics of SMPTE HD is that while it specifies performance there’s no standard for colour. Instead for those wanting to get a handle on colour there are 8-bit values giving 17 million possible pixel colours or 10-bit colour values for a one billion colour palette.
Typically, 10-bit manufacturers claim a richer colour rendition and not surprisingly given the hyper engineering that goes into its cameras, Avigilon uses 10-bit and 14-bit colour. Along with this the company’s products meet and significantly exceed SMPTE HD and other HD standards and specifications.
Avigilon is a unique player in the market in many ways, having invested in the development of powerful large MP cameras and clever software to manage them effortlessly on ordinary workstations and networks. Avigilon makes HD cameras but it also makes cameras that offer performance far superior to HD in terms of resolution and that gives the company a different perspective on the HD question.
Avigilon’s Mark Hartmann says he believes HD is becoming a default standard for CCTV and he says this is not only being driven by manufacturers but also the expectations of consumers who now demand high image resolution and detail. But Hartmann has certain reservations.
“A standard like HD is not necessarily required for all environments, just as HD does not satisfy the detail required in others,” he explains. “A careful assessment of the desired resolution of a scene must first be understood before the appropriate camera (HD or other) can be specified.”
One of the challenges of HD is that it contains a variable – the term HD does not always mean the same thing.
“HD defines a standard which in CCTV is a hugely variable,” says Hartmann. “To state that an HD camera will give you fantastic detail could be as true as it could be false. For instance, an HD camera with a lens covering a 90-degree scene will not give fantastic detail of a subject 30m away, yet with the appropriate lens it could deliver an image of the same subject in great detail.
“So I don’t feel that the term HD offers much more to end users than to suggest that the camera has a greater resolution than analog. It’s difficult for end users to visualise the difference.
“Given that the term HD typically refers to all image resolutions above analog (extending in the Avigilon case to 16MP) it is likely too early for a suggestion that HD is striking a balance,” Hartmann says.
“We are in the midst of a sensor and compression technology evolution, compounded with greater storage capacities and network speeds being offered at prices which reduce daily – thus removing the traditional (image quality, low light performance, and storage and bandwidth) barriers.
“This is allowing camera manufacturers the opportunity to question traditional CCTV design logic and present the industry with the solutions previously unthought-of.”
Hartmann agrees HD is significantly superior to SDTV in terms of performance but he says there’s a caveat.
“One should always remember that other elements (such as lenses, sensors and compression) play a major role, and in an extreme case a poor quality HD camera can look much worse than an average SDTV model.”
Something else that’s important with HD is compression – it impacts processing, storage and transmission. With bigger systems incorporating hundreds of cameras, compression becomes more important still.
“If not the most important component, compression is at least one of the major considerations in HD CCTV,” explains Hartmann. “Avigilon use JPEG2000 for the HD and Pro range of cameras, and we also offer 1MP and 2MP H.264 cameras. H.264 is widely used in the industry and regarded by most as the leading standard, however it falls significantly short of JPEG2000 when dealing with HD footage.
“When designing our system, Avigilon had a choice of compression technologies. JPEG2000 was a clear winner for its many advantages, most notably the ability to handle 16MP image data. Unfortunately it meant that Avigilon had to re-educate the market with respect to this less-known technology, but H.264 simply could not deliver on performance and image quality the same way as JPEG2000.
“Now, almost 4 years after our first installation of 16MP cameras in Australia, the industry is slowly realising the truth in our conviction – as still there is no competing 16MP or greater camera on the market using H.264,” says Hartmann.
“At present our Pro range can only connect to the Avigilon NVMS due to the compression technology which we have leveraged. This has been seen by some as proprietary, however, it is actually a situation in which other manufacturers have not adopted the JPEG2000 format and therefore have not engineered their systems to handle the requirements of a 16MP camera.”
“To state that an HD camera will give you fantastic detail could be as true as it could be false. For instance, an HD camera with a lens covering a 90-degree scene will not give fantastic detail of a subject 30m away, yet with the appropriate lens it could deliver an image of the same subject in great detail”
Another key consideration of HD is lenses. Megapixel cameras, which include HD cameras, demand superior lenses in order to perform at their best.
“Our CMOS range of HD cameras require standard C/CS mount CCTV lenses – but these should be MP rated. Hard to source and lens manufacturers are slowly catching-up, I think,” Hartmann says.
“Meanwhile, our Pro range of cameras uses standard Canon EF mount lenses designed for photographic SLR cameras. These are readily available from any camera store, inexpensive (compared to some CCTV lenses) and offer superb image quality well beyond 16MP.”
From an installer’s perspective, are there additional demands on an installer when HD is being installed compared to standard resolution VGA IP cameras? Not according to Hartmann.
“The only additional request is that they focus the cameras properly,” he says. “Given the increased resolution, HD is less forgiving when lenses are not focussed correctly.”
At this point in the development of digital surveillance cameras, HD provides a nice balance between resolution and quality but there are applications for bigger megapixel resolutions from 5-20MP in larger spaces like public surveillance, stadiums and airport terminals. It’s in this area that Avigilon’s big boys, including its flagship 16MP camera, come into their own. According to Hartmann, it comes down to a focus on selecting the correct camera for the correct environment.
“Just as it would be wrong to place a 16MP camera in a doorway, it would be wrong to put a 2MP (HD) camera in a stadium,” he says. “For this reason Avigilon use a standard unit of measurement (Pixels per Meter) to assist the customer with selecting the best cameras/lens combination for a scene.
“We have created a tool which accepts three parameters (height above target, distance to target and scene width) and calculates the pix/m value. After showing the customer an image depicting the resulting value we can then select the correct camera and lens for the scene.
“Naturally a 16MP camera does a fantastic job in large areas while offering benefits such as a reduction in camera numbers. Avigilon recently sold 12 x 16MP cameras to a large 52,500 seat stadium. With these 12 cameras, we will capture every face in the seating bowl – and never miss an incident!”
Video Security Products distributes Sony’s complete range of progressive scan HD solutions which includes HD(720p)and full HD(1080p) models up to 3MP with most models supporting DEPA (Video Analytics). Sony also has single and multi channel encoders, hybrid network recorders, and Sony’s intelligent monitoring software, Sony RealShot.
Along with this strong range, Video Security Products also stocks and supports VMS software, NVR and hybrid NVR’s from ExacqVision.
When it comes to HD, the company’s Zaki Wazir says he thinks HD is becoming a standard because manufacturers are putting their marketing efforts behind HD CCTV. Wazir feels the HD standard makes life easier for end users struggling with choice.
“I believe as a standard it does make life easier because there are fewer variables,” he explains. “With HD, there is no choice of aspect ratio, frame rate is set, and only two video formats (1280×720 and 1920×1080). HD also has a quality compliance standard.”
Wazir says in his opinion HD is the right balance between image quality, low light performance, and storage and bandwidth demands.
“HD is a good balance, however it may not a clear case of being better. In some situations greater resolution may be required than what HD can offer,” he explains.
“Compared to SD cameras HD looks absolutely awesome – there is no comparison to SDTV. HD gives you the full 16:9 aspect ratio and a high frame rate. Other advantages when using Sony HD, for example, is that some models offer Sony’s “Stream Squared” function which allows for the streaming of two 4:3 aspect ratio video in SD resolution which allows you to replace two SD cameras with a single HD camera.”
Wazir believes that HD is an end-to-end solution and he feels all components of a system must be able to support HD – from comms to switching to storage. Compression too, is an area of real importance.
“These are critical parts of the overall solution and quite often overlooked,” he explains. “Switching, for example, must be capable of not just the power requirements of PoE, but also the throughput required.”
“And when you offer the resolution and frame rate that HD offers, if you don’t have efficient compression it will all just bottle neck.
Sony cameras offer H.264, MPEG4, and JPEG compressions to handle every application.”
When it comes to lenses, Wazir says HD cameras require megapixel lenses.
“I believe that all HD cameras are megapixel, but not all MP cameras can offer the HD standard,” he explains. “And I don’t believe that HD cameras require anything other than MP lenses.
“As far as we are concerned, Sony supplies lenses with all their HD cameras so it will never be an issue for our customers but with other manufacturers it might be.”
Wazir says another advantage of Sony’s universal recognition is that the company’s HD products can be supported by all video management systems.
“Sony being the most recognised name in CCTV, I can’t think of one that can’t integrate with our products,’ he explains.
“Nor are there challenges for installers using Sony product – just make sure you use quality cable, switching, and storage hardware and use the applicable bandwidth calculators. If fact, thanks to the SONY Easy-Focus button, Sony cameras do everything but the physical installation.”
Where bigger spaces are concerned, Wazir agrees there are applications where larger megapixel cameras than 720p or 1080p offer superior performance.
“In certain applications, yes I would have to agree bigger megapixel cameras have a role,” he says. “HD after all is definition 1920 x 1800 pixels or 2.1MP.”
Meanwhile, the Integrated Products’ IndigoVision range provides a end-to-end IndigoVision solution managed by IndigoVision’s Control Center complete management software through to cameras (Fixed and PTZ), encoders, linux based hardware NVRs and Windows NVR software for use on a PC platform.
Integrated Products’ Stuart Fowler says he believes HD is the standard for higher resolution cameras.
“Yes, we certainly believe HD is the standard for higher res. Megapixel allows manufacturers to go to whatever higher resolutions they want, but these are proprietary,” he explains.
“The true HD standard is well defined, at 720p and 1080p with full frame rate. Megapixel, on the other hand, is just a loose term to mean greater pixels than SD at any aspect ratio. HD occupies the sweet spot to which CCTV will move from SD over the coming years.
“Our view is that there is a place for higher resolution cameras than HD but these are niche offerings which won’t see volume. Eventually, a long time into the future, we might see a successor to the HD standards which takes resolution above 1080p…but HD will be the mainstream of the CCTV market for many years to come.
720p best tradeoff
“At the moment I think HD 720p is probably the best trade off between all these aspects. Even today, we want to fully support SD for low light performance and cost-effective bandwidth and storage and we allow people to seamlessly mix SD & HD within a system.”
When it comes to colour palette, Fowler says IndigoVision employs both 8 and 10-bit technologies.
“We have a range of HD cameras which include 8-bit colour and 10-bit colour,” he explains. “We use the best sensor technology available for our cameras for image quality and low light performance regardless of which colour palette used. We see very little difference in the final image between the two palettes as there are many other variables which effect image quality.
“Our current products conform to SMPTE 720p resolution requirements and we are developing products for SMPTE 1080p resolution. One slight deviation to the true SMPTE standard is frame rate – we use 25p/30p instead of the 24p/30p defined in the standard as certain verticals require full frame rate video at all times. Casinos are an example of a vertical in which full frame 25/30 is required, and 24p doesn’t make the grade.”
“With HD, there is no choice of aspect ratio, frame rate is set, and only two video formats (1280×720 and 1920×1080). HD also has a quality compliance standard. HD is a good balance, however, it may not a clear case of being better”
Fowler is circumspect when it comes to just how much better HD is compared to SDTV from the point of view of raw performance.
“720p has theoretically 3 times the pixels of the maximum standard definition,” he explains. “But in reality, raw resolution is somewhat meaningless and what matters is how well the video is compressed. You can have SD that gives better video quality than HD, if the HD compression is poor – this means one HD manufacturer’s 720p can be better than another manufacturer’s 1080p.”
“Compression is an important element of any HD solution – raw pixel data is going up by 3-10 times when you move from SD to HD. Lots of IT managers were already unhappy about SD bandwidths taking space on their network, so compression becomes even more important with HD.
“And obviously, any high resolution camera requires good lenses; if you put a poor lens on an HD or a megapixel sensor you’ll get poor video. In general, lenses are struggling to keep up with sensor resolutions.”
When it comes to managing HD, Fowler says there are some things installers need to take into account.
“Focusing is paramount,” he explains. “Bandwidth is higher, so network backbone and storage needs to be well speced. Also, workstations can struggle to decode many HD streams in parallel, so this needs care also.”
At Pacific Communications Kieron McDonough believes that HD is not the standard right now but he says higher definition cameras will become the standard in the near future and will encourage migration to IP-based cameras.
What is HD?
“HD is still a big variable,” he explains. “The question is, ‘What is HD?’ I think terms like Megapixel, 720p, 1080I/P, SMPTE and even the term itself ‘High Definition’ just confuses people as it gets talked about without clarification. Let alone mentioning other non SMPTE resolutions that are of higher resolution.
“In simple terms, the SMPTE standard quantifies the specifications but this is only a subset of high definition. Most people in the industry would understand the term ‘High Definition’ to be greater than the traditional PAL/NTSC resolution (equated to a digitised 720 x 576 pixels), which works out to be 0.4 Megapixels.
“So in real terms, anything that is around 1 million pixels or more is considered or generally understood to be of a ‘High Definition’. This said, a lot of our cameras meet these standards and some of these camera vendors go above the SMPTE standard to 3, 5, 7, 10 Megapixel and higher. But the question is, will SMPTE keep up? Does it need to?
“We sell most major vendor cameras and we do our own in house testing to compare performance. We sell a wide range of vendors’ cameras, catering for virtually all requirements. It is important to note though, you can have a great quality camera but the lens and the monitor are just as important to reflect the image accurately.”
“At the moment I think HD 720p is probably the best trade off between all these aspects. Even today, we want to fully support SD for low light performance and cost-effective bandwidth and storage and we allow people to seamlessly mix SD & HD within a system”
According to McDonough one camera can’t do it all.
“There are general cameras and specific cameras. This applies to both SD (Standard Definition) and HD cameras. As technology continues to improve, there will be better low light HD cameras, making them versatile and suitable for wider applications. In terms of storage and bandwidth, it really comes down to what is required and what the customer is willing to spend. HD does come with its own considerations.
“Without a doubt, HD is one main reason to go to IP cameras. When you put one up against the other (SD), the difference in picture quality is amazing. Cast your mind back to traditional television analogue picture to that of today’s HD picture. I believe, once you have converted, you will never go back.
McDonough believes it’s important that all components of a system are able to support HD – from comms to switching to storage. Compression too, is an issue end users and installers must consider.
“This is vital without a shadow of a doubt,” he says. “The system is as only as good as the weakest link – it’s a cliché but it is 100 per cent true.
“When it comes to compression there are several common formats. The industry standards, MPEG4 (MPEG4-part 2), H.264 (MPEG4-part 10) and MJPEG are the most common,” he explains. “There are others that are not so common such as JPEG2000 and proprietary variations.
“Clearly, the MPEG4 series are great and the H.264 technology is even better. The biggest advantage of H.264 is the fact that it requires less storage than the others for similar resolution and frame rate. Network consideration is also important and while Gigabit Ethernet is becoming quite common, smart network utilisation is essential and H.264 helps in this area.”
Importantly, McDonough sees lenses as a key issue in HD, with the needs being the same as those of megapixel cameras.
“When it comes to lenses, since HD it is in the megapixel range of resolutions it needs megapixel quality lenses. As I have mentioned earlier, from the lens to the monitor, including network and storage, all need to be considered to make the HD solution work well.
“It is also important to note that, the area of interest is not always in the centre of the image and hence the lens needs to be in focus in the outer areas of the scene in question. The GBO Infinity Lenses are ideal for this and we are happy to have this great range available to our customers.”
In terms of additional demands on installers, McDonough says these mainly relate to design considerations.
“This covers lenses, cabling, storage, server and workstation needs and the monitor,” he says. “From a simple physical installation aspect, very little, but the infrastructure is the consideration.
“When it comes to higher megapixel counts, there are projects that will require very specific needs. Extremely high definition is one of them, like thermal cameras. We carry a wide range of cameras to suit any customer’s needs.
“We stock most of the major brands of HD cameras and a number of IP recording solutions. Our best suggestion is for the customer to come in to one of our branches, sit down and we can demonstrate the differences and let them pick the solution best suited for their application. The customer can also bring in their client (end user) and use our facilities to educate the end user.”
At Bosch Security Systems, Phil Brewer says HD offers a performance compromise between SD and larger MP cameras, but he says HD, with CCD technology, offers better low light performance, improved dynamic range and less bandwidth than MP.
“In particular, bandwidth is a crucial factor for IP CCTV systems,” Brewer explains. “HD is an evolutionary step for users of SD, so it can be seen as a logical evolution to an industry performance standard.
But Brewer says the fundamentals of video surveillance need to be taken into account regardless of the nature of the camera and higher resolution cameras offer challenges of their own that must be taken into account.
“Any camera, whether it is SD, HD or MP, needs a suitable amount of light on the scene. For SD the amount of light is quite low compared to HD or MP; this is due to the relatively small pixels in HD and MP sensors compared to SD sensors. There is no getting around this if CCTV systems continue to use 1/3” and ½” format lenses – which in mainstream CCTV applications this will be true.
“If there is enough light on the scene when the system needs to be used, whether it is public surveillance, stadiums or airport terminals, then we will see 5-20 MP cameras being used. But, with higher energy costs likely to affect infrastructure owners in the short to medium term, the cost to illuminate such venues may become cost prohibitive with existing lighting technologies such as fluorescent lamps, metal halide flood lamps etc, to make use of the 5-20 MP cameras.
“Lighting for security applications is already an area that Bosch has addressed with its range of LED based white light and infrared (IR) technologies. They are especially designed for the security market to provide constant light on the scene of interest, without the light degradation that is typical over the lifetime of bulb, lamp or gas-discharge lighting technologies.
“Finally, LED based technologies have a much lower current draw, which means they use less power – all the more reason to explore LED based lighting if/when a carbon tax or credit scheme is introduced to all industries. Light – it is the most important element in the CCTV system.
“HD should make life easier for end users, IT specialists, consultants, etc because it is so easily defined. If it meets SMPTE 296M-2001 (i.e. 1280×720 pixels, 16:9 aspect ratio, etc.) or SMPTE 274M-2008 (i.e. 1920×1080 pixels, 16:9 aspect ratio, etc.) then it is a HD camera; if not, then it is not an HD camera. The entire Bosch range of 720p and 1080p cameras conform to the relevant SMPTE standards.
“However, there are several manufacturers that call their products ‘HD compliant’ but do not produce either of these two resolutions. That must be very difficult for end users and other members associated with, or entering, the CCTV industry to understand.
When it comes to HD offering the right balance between image quality, low light performance, and storage and bandwidth demands, Brewer offers a thoughtful reply.
“For general applications the answer is yes; but for harsh scenes with low light, strong backlight conditions or other difficult lighting situations, high performance standard definition cameras still have the advantage of producing a useful picture with the correct placement and field of view,” he explains.
“Without a doubt, HD is one main reason to go to IP cameras. When you put one up against the other (SD), the difference in picture quality is amazing. Cast your mind back to traditional television analogue picture to that of today’s HD picture. I believe, once you have converted, you will never go back”
“For applications that are based on security, usually the prime time that CCTV is required is during these difficult lighting conditions and that must be considered in a system design. Image performance is still paramount, no matter if you are using SD, HD or MP cameras.
“I would say that in terms of raw performance, i.e. pixel count, HD is a big step in the right direction. HD delivers additional resolution that is simply not available to SD cameras that were ruled by the PAL (Phase Alternating Line) and NTSC (National Television System Committee, or sometimes Not Twice Same Colour by its detractors) standards. Essentially, PAL was limited to 576i and NTSC was limited to 480i. The HD standard also unifies the PAL versus NTSC formats; HD is independent of either PAL or NTSC.”
And Brewer says image elements like colour must also be considered very carefully.
“Bosch HD cameras comply with ITU-R BT.709 for colour representation, and refers to HDTV systems having about two million luma samples per image,” he says. “It also specified the number of images per second, i.e. 24, 25, 30, 50 or 60. Claiming 8-bit or 10-bit is somewhat misleading because that may be the capability of the camera, but what about the graphics card, type of monitor and software on the workstation?
“All of these are linked to the colour representation and usually treat an encoded video signal in the ‘PC world’ which is different to the ‘video world’. For example, does the video card treat black as 0 and white as 255? Or does it reserve 1-15 for ‘video footroom’ and reserve 236-254 for headroom while using 16 as the black level and 235 as the white level?”
When it comes to end to end HD capability, Brewer says this is vital from the point of view of future-proofing the end user’s system for future expansion.
“As technologies improve, such as more HD cameras supporting the ONVIF standard, the end users’ infrastructure needs to be ready to support those new technologies,” he explains.
“In the past with analogue video systems, an RG-59 coaxial cable was future proof and ready to accept any new analogue cameras that were released to the market. With HD, that using the IP/IT infrastructure, the capacity of that IP/IT infrastructure needs to be taken into account at its installation but also for the useful lifetime of the facility, which is typically greater than 25 years.
“Compression is very important for HD to reduce the bandwidth and storage requirements, because a HD image has more information in it compared to SD. Uncompressed SD (PAL) is about 160Mbps, but uncompressed HD (720p) is about 840Mbps and uncompressed HD (1080p) is about 1500Mbps. These data rates are huge compared to the capability of a 1Gbps network.
“Bosch uses H.264 compression to achieve the most of compression algorithms, but also pairs that with a suitable processor with enough computational power to produce 1080p30 or 720p60 video streams. This is another requirement that some other manufacturers’ fail to produce a true HD video stream as they can only manage 720p20 – that’s only 20 images per second in HD.
“Bosch HD cameras are also able to generate four video streams (quad-streaming) at the same time to send the video stream to several devices/users at the same time; again, I invite you to compare with other manufacturers’ that claim H.264 but can only produce one or two H.264 streams.”
As a manufacturer, Bosch offers a complete product portfolio of analogue and IP cameras, encoders, DVR/NVR’s etc.
“These range from harsh environment pan-tilt-zoom (PTZ) cameras to internal HD dome cameras, management software to iSCSI storage arrays with RAID-DP redundancy,” Brewer says.
“The portfolio is managed by Bosch in Germany, with Bosch R&D and factories producing the products. All products have a 3-year warranty, and are well known for their proven product reliability. Bosch Security Systems in Australia is a wholly own entity of Robert Bosch GmbH, which gives us product support advantages that wholesalers and distributors
of other manufacturers’ do not enjoy.
“All Bosch HD cameras are progressive scan, even the Dinion 720p and FlexiDome 720p HD cameras that use a CCD sensor,” Brewer explains.
“Bosch is able to control the CCD in a progressive mode. While there is always talk about how much CMOS sensors have improved in low light performance over the years, and that is true, what they neglect to mention is that CCD sensors have also improved over the years.
“Think about the low-light performance of colour CCD cameras from the mid-1990’s compared to colour CCD cameras that people use today. The CCD sensors in Bosch’s Dinion 720p and FlexiDome 720p cameras are especially designed to Bosch’s specifications and are unique in the market for low-light performance for HD cameras.”
A nice quality of Bosch HD cameras is that because they are ONVIF compliant they can easily be integrated with third party management solutions.
“ONVIF is a standard that allows third party management systems to connect to ONVIF compliant cameras and encoders,” Brewer says. “ONVIF still has some way to go to allow storage and video analytics to be shared between different manufacturers. However, with the announcement and release of ONVIF 2.0 in November 2010, manufacturers are working towards a complete IP/IT system that is independent of manufacturers Video Software Development Kits (VSDK’s).
“At the moment, to have all functionality of any IP camera or encoder, the manufacturer of the management software needs to use the IP camera/encoders manufacturer’s VSDK and write interfacing software. This can be a problem for the management software manufacturer because they need to write software, test it and finally release it to the market place.
“For the IP camera or encoder manufacturer, there is frustration that their product has been released but they are waiting for third-party management software manufacturers to catch up. In the end, it means that the end user, may be forced to use hardware or software that is less than ideal for their application because of they are waiting for VSDK’s to be implemented. ONVIF will eventually fully address this issue for network CCTV systems, including HD CCTV.”
Brewer says there are some additional demands required of techs when installing HD cameras compared to VGA IP cameras with respect to ensuring that the back focus is done right.
“The installer will need to check that the network infrastructure will be able to cope with the additional data from a HD camera compared to a VGA IP camera, and this includes that the storage medium will be able to record long enough at the required resolution,” he says.
“Finally, a decent laptop with a highly specified graphics processing unit (GPU) is a must for either demonstrating or commissioning HD cameras.”
“Uncompressed SD (PAL) is about 160Mbps, but uncompressed HD (720p) is about 840Mbps and uncompressed HD (1080p) is about 1500Mbps. These data rates are huge compared to the capability of a 1Gbps network”