IP-based video envisions a brighter future

The video surveillance field appears to some outsiders to be embracing wholeheartedly the shift from analog to Internet Protocol-based digital cameras, but to insiders the transition has been at times frustratingly slow. To get a handle on the pros and cons of analog vs. IP-based video surveillance systems, GSN's editor-in-chief, Jacob Goodwin, recently spoke with four experts in these emerging IP-based technologies:

Fredrik Nilsson, general manager of Axis Communications AB, a supplier of IP-based cameras and equipment;
Marc Beningson, director of transportation security systems for systems integrator, 4D Security Solutions, Inc.;
Robert Beliles, senior physical security marketing manager for Cisco Systems, Inc., the giant manufacturer of routers and other Internet-based infrastructure;
Dan Rittman, director of engineering at Exacq Technologies, Inc., a supplier of advanced IP-based video solutions.

Why is it taking the video surveillance field so long to shift from analog to IP-based cameras?

Looking at the market, it's pretty amazing that 90 percent of the cameras still being installed are analog cameras, given that IP-based systems are better image quality, more scalable and many times lower cost. The main reason is that the incumbent players in the market - consultant system integrators and existing vendors - are reluctant to change to this market because it provides more competition for them in an unknown area.

In my field, which is transportation security, there is already a lot of transition to digital video, but we're still using analog cameras and encoding them at the edge because we're using a lot of highly specialized cameras that have to work outside - thermals, near-infrared and other specialty cameras. In those areas, IP cameras simply aren't there yet.

A lot of the challenge is that IP cameras haven't been around as long, so there's a certain desire to make sure that the technology's there. The industry tends to be fairly risk-averse, in terms of their selection of products. There's a certain amount of vetting of the existing products that are out there, some of which are quite good. Where IP cameras really excel is in their ability to have power over the Ethernet, which simplifies the cabling infrastructure and the maintenance. IP cameras can go far beyond the VGA resolutions and do some pretty fabulous things, such as not necessarily requiring a pan/tilt/zoom camera in order to have a much broader field of view.

Part of it is an education process, so the customers understand what IP cameras really bring to the market and proving that they have been vetted. More and more customers' successes will feed upon each other and you'll see that adoption rate increase dramatically over the next two to three years.

One thing that people tend to forget is that seven years ago there weren't even DVR's (Digital Video Recorders), so the security industry went through a major change just seven years ago shifting from VCR's to DVR's. Dealers are just assimilating all that. To throw another substantial change at them this quickly is difficult. They're just now getting their hands around all of the DVR complexity.

What are some of the primary capabilities that exist with IP-based surveillance systems that aren't available with analog systems?

The most important thing with any video surveillance system is image quality. There are two things you can do in network cameras today: progressive scan, which provides much less motion blur when you have movement in the picture, and megapixel resolution. Also, video transport cables can be used for more than just a video transport. The Ethernet can also provide the power. You have audio capabilities and pan/tilt/zoom control capabilities all running over the same cable.

That provides two things: a measure of cost savings, which is important in any installation, and the ability to do backup easily from the server into each of the cameras, so in case the power goes down in the facility, you can still have the security system in operation. That has been very difficult and very expensive to do with analog technology.

Everybody opens their paper every Sunday and sees ads from Circuit City or Best Buy for a 10 megapixel camera. They say, "I want that." Then, they go and look at their security system and they've got only a 380k pixel camera out there. That's what's going to drive it - the high resolution. It lets you do so much. For example, customers have used the high res camera for a wide field of view and still been able to digitally zoom in to track labels on packages leaving their facility. That's the sort of thing that's really going to drive the market.

To be fair to the analog world, what are the technical shortcomings or disadvantages of shifting to IP-based systems?

What becomes interesting is the ability for migration technology - such as encoders - to provide the capability to gracefully migrate to an IP world. You don't necessarily have to go up with a cherry picker and replace the existing camera that is working just fine and dandy for a particular application. Now, with an encoder, you have the ability to be able to keep that analog camera in place, digitize and compress that video and send it over the IP network, basically providing a lot of the benefits you would get with an IP camera while continuing to leverage your existing investment in your analog infrastructure. We see that as a big driver in the adoption of IP cameras.

One of the difficulties is integration. Coming out of the back of every analog camera in the United States is NTSC video. We know what that is. You can play with it. You can ship it over fiber. You can convert it. You can feed the analytics. But, if you've got IP, you've got integration and you've got to make sure that everything works with everything else.

Your traditional security contractor has been working with serial interfaces and RS-485 for 25 years. He's not a guy who's going to have a software whiz or an IT whiz in the back room who's going to be able to make this all work together.

What are the unique concerns of government customers when it comes to IP-based camera systems?

The issue is one we've heard a lot of times, particularly if there's any inter-agency cooperation or the requirement to share video amongst different groups. Toy have the challenge of potentially having to share when a lot of government organizations have a culture that says, "Don't do anything with any other group." That tends to be one of the biggest challenges that the industry faces today.

When it comes to government, we've seen a couple of drivers. One is cost efficiency because government business always goes out to bid to make sure they get the best solution at the lowest price. The other one is being able to have secure communication. If you look at analog video today, that is not truly secure. Anyone who tape into that wire can see the video.

With encrypted video from a network IP-based camera, you will not be able to do that. Also, when it comes to storing video for many months and being able to have redundancy and backup, that's something the IT industry has been addressing for many years.

One thing that's going to be more widespread in the coming year is video analytics of all sorts. People have seen the movies with facial recognition - these things have just been pipe dreams - but as we go around and talk to people, these analytics are coming into their own. The processors have gotten fast enough for us to push this stuff out to the edge of the network. Analytics is really going to be a major player within the next two years.

I agree. Analytics is maturing. We're going to see a lot more analytics out on the edge, where it belongs, as opposed to back in the server room. We'll see a lot more plug-and-play, where I could take his camera and his camera and his analytics and your switch and make it all work without needing a full IT and software development department.

Video analytics will facilitate a level of access control that isn't currently available today. Once you have that capability, you'll be able to tie physical presence into network access, and tie it into fire alarm systems and a lot of other systems. It's that level of integration, where you're going to start seeing video playing a role outside of its normal observation mode.

Another thing that's starting to happen is fusion - taking data from a variety of sources, like analytics, radars, and other kinds of sensors, and combining them to increase the probability of detection and to reduce the false and nuisance alarms, both inside a building and especially out on the perimeter.

What do you see happening to the marketplace for analog products versus IP-based digital products? Where would you lay your bets if you were betting on this industry?

I am betting on this industry. We're a small start-up company and I've got a substantial bet that it's going the IP way. And that Exacq may buy Cisco in the future.

We've seen multiple waves of convergence. We've seen mainframe computer communications. We've seen telephony. They've converged onto the IP network. There are a number of vendors who have already done the job to get it started. Cisco is now in the industry. We're going to articulate how you can successfully deploy physical security over the IP network.

There's no question that IP is the way of the future. The question is how far into the future and how long is this transition. As long as there's something like a terminal server, the serial world is never going to go away, and there may be no reason for them to go away. But the core of security systems will definitely be IP-based. Just as we're talking about IP cameras today, a lot of other security sensors will be IP within one or two or three years.

We launched the first network video product in 1996. That's 10 years today, so it's been a long time coming in the market. Maybe that has made some of the analog players say that (the IP niche) is still small, it's growing slowly and we can catch on. But, I think it will provide much bigger change than people envision today. It's very difficult for a lot of the incumbent players to change tp these new market dynamics, to the new technology, to the new sales model and to the new partnership model that are evolving from the shift to network video technology.

It's always interesting to see what happened to other industries when you had a big enough change. Look at digital photography, for example, and see what that did to Kodak.

The paradigms are shifting totally. Not all, but some of the dinosaurs will disappear.