Whats The Future of CCTV
by Security Electronics and Networks | @Analysis Articles | June 24, 2013, 7:00am AEST
Q: Vlado, I can’t help feeling we’re at a cross roads for CCTV. Some of the things on the horizon look threatening, others look positive – do you think the current model of CCTV distribution and installation has a bright or an uncertain future?
A: I agree John, from the point of view of technology, we are at the crossroads, but I don’t think the uncertainty is because of new technology, but rather because of aggressive marketing from low cost, low quality manufacturers. Some small businesses think they can make a quick buck and huge profit by buying direct and re-selling low quality questionable copies, but in the long term, we are all losing because our industry is only so small. This price-driven low quality marketing by mostly Chinese manufacturers is the actual problem I see. Somebody could make a quick buck on the way, but the industry will lose its reputation in the long run.
Q: Is the market’s focus on price likely to have a negative impact in the medium and long term or is this just part of the normal cycle?
A: To be fair, Australian customers always had price-driven attitude. Nothing against price-driven attitudes per se, but we shouldn’t make the price the decisive factor, but one of the factors. Quality and innovation should be most important. I always believed, and I still believe, with better education, customers always choose the better product, irrespective of the price. When you think of an HD camera as just another HD camera and put it in the same category as all other HD cameras on the market, irrespective of its quality, then you have a problem.
So, yes, prices are going down, but they can’t go any further than the minimum cost. Once you buy such a product, you lose respect and appreciation for it. If it breaks down, nobody repairs it anymore, but simply throws it out and buy a new one. Second time, however, usually one thinks of buying a bit better and more expensive, so you don’t have to throw the product away so soon. Eventually we learn that cheap is expensive in the long run.
Q: Would you agree that the price falls in HD, which must lead to falling margins and build quality or both, will push serious manufacturers in pursuit of the next technological edge?
A: Yes, I agree with that, but I think the next technological edge may not necessarily be more pixels (4k instead of HD) but rather more intelligence in image processing, low light performance, automatic detection of events, or similar. Certainly, sensors with more pixels will become a reality too, but it will be a few years until then.
Q: Some threats could include generic cameras with Win7 operating systems and the headlong rush towards the tiny and the low cost in HD video cameras. The simplification of HD seems to me to have happened very fast – in just a couple of years. Does it seem that way to you?
A: Honestly, HD cameras with Windows7? I hope the Windows firewall will allow the camera to operate! It will ask: “Do you really want this application to run? Are you sure?” For me the primary challenges are that when there’s progress and innovation somewhere in the world, that company decides to make a new product. Then, in order to make it more affordable they choose to make it in China.
But they forget that there is a back-door in the same factory where, despite the copyright agreement, the product is sold to others at even lower prices by slightly modifying the design so it looks different. I think this is happening with the low cost HD cameras you mentioned. It is a fact that the cost of image sensors manufacturing has gone down as well, but it is the copying and cutting corners in quality that is damaging.
Q: There are many things that are positive in CCTV, including the development of 4K monitor screens, the development of 4K CCTV cameras – Bosch is first cab off that rank – as well as the ratification of H.265, which will make 4K useable on typical surveillance subnets.
Something else that’s exciting is Samsung’s 1Gbps 5G technology which will allow an HD movie to download in one second. That’s a technology the surveillance industry could learn to love. What other developments excite you?
A: Specifically in CCTV (and maybe I am being biased here) the Dallmeier’s Panomera is a fantastic concept that I believe, even if it is jaw-dropping now as it is, it will get even better as the faster computers and more capable software become available.
In CCTV technology in general, I think the next big thing would be certainly 4k sensors, but the display will play an important role too. At this stage there are not that many super-sharp screens with 8MP displays, but I believe soon OLED technology will come out of the labs and become a real and more affordable product. This will offer incredible possibilities, especially for control rooms.
I can imagine a wide and curved OLDE display in front of an operator’s desk that will show all video signals one can select, scale them up or down any which way one wants, all on a thin and transparent OLED film, and of course running on Linux or Mac.
Q: IR illumination. This is a great development in my opinion. Do you think it allows manufacturers to focus the power of their chipsets on
areas of performance like WDR and improved resolution?
A: No. In order for any sensor, be that CCD or CMOS, to see colour, they must have IR cut filter, because sensors are sensitive to IR light more than visible light. So, if you have IR cut filter you can see colours properly, but you can't see infra red. If you want to see infra red you have to remove the IR cut filter, but then you won't be able to see colours.
An additional side-effect is that IR light has longer wavelengths, lower frequency than visible light, and as such penetrates deeper but also the focus point is a little bit behind the sensor plane (hence you need an IR-corrected lens).
If the sensor is designed with larger pixels (meaning a bigger sensor size) it will automatically be more light sensitive. But, as we know, increasing the sensor size increases the cost, as well as requiring bigger and more expensive lenses.
Furthermore, there is not really much processing time to be spent on the sensor itself, but rather this is done in the electronics after the sensor. The current trend is a little bit of image processing, but not really on the sensor itself but in the post-sensor electronics, like double-exposure.
My belief is that there will be a new sensor design that will stay small but increase low light capability in some other clever way, one of which is by back-illumination. This is already done on the iPhone, for example. The chip is exposed from behind the pixel area, which is less obscured by conductor micro tracks, which are on the top layer of a sensor.
Q: We’re seeing a significant broadening of the market for hemispheric cameras. Do you think we could see ubiquity with this flexible type of camera? Or is this horses for courses?
A: I think horses for courses. It all depends on how it is done. Mobile phones were mobile phones until iPhone came out. Nokia, Motorola, Ericsson and Blueberry were the leaders and nobody thought it was possible to do anything radically different and new, but that changed with the iPhone because some people spend a lot of time thinking more
creatively. And now everybody is copying it.
Perhaps hemispheric or immersive technology cameras may become more popular if they are designed to offer the best of all that they are good at, rather than just an idea of hemispheric viewing. I personally don’t find them that attractive today, but that doesn’t mean they won’t become much more attractive if resolution is improved, low light capability, and there are improvements to the lenses they are using.
Q: Do you see a way that cloud can have a genuine business model outside of smaller commercial and domestic applications? Given the huge drop in cabled 1G WAN links in city centres – they are about $600 a month in Australia now – I think it’s increasingly possible to employ a cloud model there’s just no one seriously pushing it. Would you agree?
A: I thought the same John, at the introduction of the Cloud concept a few years ago, but I think that hasn’t happened yet because of the lack of feeling for the storage media presence. We all have now our photos, music, e-mails and DropBox files in the cloud, but somehow I still want to have my own copies somewhere on my physical drives near me.
Also, I know that a lot of security managers will not allow their data to go somewhere uncontrolled (despite the encryption). Plus, the NBN is a long way from being complete, which could be another reason for the lack of popularity. I do have in my place, believe it or not, a 100 Mb/s internet, but I find it limiting by the other side – the server I am talking to.
If they don’t have the pipe coming to me, I can’t see anything more exciting at 100 Mb/s. But I am sure, these are the things that will get better each day, it will just take time. Maybe then, when
all preconditions are met, the cloud surveillance may be a bit more attractive possibility.
Q: Analytics – for a very long time we’ve all been happy enough with our triplines and now we have BRS suggesting its AISight reason-based analytic is the perfect solution. I like AISight but it obviously prices in development and that means SME and cash strapped users can’t afford it despite the technology’s discernment. Where would you like to see analytics go in the future?
A: Yes, it has been painful to see a little bit of decline in the video analytic trust in the past few years. I think this is because some people , especially the sales, over-hyped it. Still, this is a very
promising area, where some good products are already available in the market, but the important is to note that no video analytics is as perfect as human eye and brain.
As long as we learn to expect a realistic outcome, video analytics can be very useful in reducing the stress on the operator watching hundreds of cameras, and take his/her attention to a potential incident.
Counting, numberplate recognition, face identification, are all existing products, some better than others, but none perfect or 100 per cent accurate as there are too many non-predictive variables. When a customer understands this, and accepts it, the analytics becomes a very useful tool, in its expected tolerances.
The good part is there are quite a few Australian smart developers that have made, and are making as we speak, fantastic products. All they need is recognition and push by the industry. With such an encouragement and a little bit of extra cash, miracles are possible. I think we as an industry should be more pro-active in supporting local products.
Q: Flash storage – we have 128GB SDXC cards and recently a 1TB USB stick was released. Does the growing power of this technology increase the ability of integrators to push storage to the edge so as to ease network load or is on-camera storage still seen as a back-up at best, in your opinion?
A: I think this is one of possibilities I would like to see, too. Recording at the camera edge might be a very clever solution providing it is protected, and works seamlessly. I am thinking here of the file formats and being able to write many times to the local solid state storage. Also, replacing a faulty USB or SD storage should be easily done. It is also possible that someone comes with a solution where wireless wi-fi on an SD card is used (already available in the photographic world).
Q: Lenses – we know they are important but it still seems to me the majority of installations go in with fairly average lenses. Is there a percentage of loss it’s possible to apply to use of poor lenses?
A: Definitely, some of this can be actually measured reasonably easy with my ViDi Labs test chart. I have numerous samples where you can quite clearly see, the same HD camera with the same compression setting but different lenses looks so different.
It is difficult to put percentage on it, as you still see a picture, but the details might be lost to the extent that if you can satisfy the face identification requirement as per the standards, with a good lens, the inferior lens will make such a camera and setup not pass the test.
Q: Standards. We have a number of standards in the wind – one coming from manufacturers and the other coming from a standards organisation in Europe. What do you think end users and installers should pay attention to when it comes to standards?
A: This question is the most important in my opinion, since we made some real progress with Standards Australia, working together with my colleague Les Simmonds. The standard organisation in Europe you are mentioning is actually IEC, the International Electrotechnical Committee, under the umbrella of ISO.
It took us over a-year-and-a-half, but we managed to complete some procedural changes and make Australia become a voting member of IEC, rather than just an observing member, which was the case so far. Some of your readers may remember, that we started working on the Digital CCTV standards some 5 years ago, after we completed the first analogue CCTV standards Australia ever gotten, the AS 4806.1, 2, 3 and 4.
We expected the industry to come back to us with contribution, changes and comments, but unfortunately not much has happened, other than what we produced with my ICU suggested concept. Working on the standards in Australia is a voluntary job, unlike overseas, and it is difficult to demand big results from an unpaid activity.
Also, Standards Australia themselves had some internal structural changes so we lost a bit of contact with them for a while. Fortunately, part of that initial work has been noticed by some overseas standards people, including the IEC members, and I am told they used some of those ideas in the new IEC 62676 digital CCTV Standards. When we found out about these developments, we couldn’t officially contribute because of the status we had as an observing member in the IEC.
This standard is nearly finished document and it will allow us to adopt it fully, or partially. This will become the new and long overdue Digital CCTV standard (which should become 4806.5) without the need for Australia to re-invent the wheel. We already started the process of creating a new committee and will start working on it. This is a great prospect for our industry and I encourage everybody interested to help.