Home > Intelligent Video > Surveillance cameras less important – relatively speaking

Surveillance cameras less important – relatively speaking

There sure is a lot of talk these days about cameras. Cameras showing up on street corners. Cameras being installed on buses and trains. Cameras in summer homes and city parks. Cameras everywhere. But the thing that surprises me is how unimportant that little camera is in the scheme of things. Surveillancecamera

When I ask a homeowner why she wants cameras at home, she replies it allows her to watch the babysitter when she’s away. When I ask the car dealership owner why he wants cameras, he says it helps him cuts costs for guards. When I ask the casino manager, he talks most enthusiastically about the ability soon to identify high rollers walking through the door.

To just about everyone I talk to, cameras are really not interesting. But the software and services associated with the cameras certainly is. I think the age of the camera has passed, and the camera itself is now mere a data collection tool which feeds information into other systems. It is these other systems which produce the value we are looking for.

As time goes on, video surveillance will draw lessening value from cameras themselves, and more from software and hardware complementing cameras. Technical issues such as increased bandwidth availability, innovative storage solutions and manufacturing breakthroughs that reduce implementation costs will outdistance new hardware technology in their impact on video surveillance markets. Developments in software control, intelligence at points of observation and improvements in backend operations of recording, storage and retrieval of video data, will also play key roles.

Categories: Intelligent Video
  1. August 15, 2008 at 10:06 am

    Hi Steve,
    This is a fascinating post. I agree, in principle, that cameras will become relatively less important than the software and hardware around it. I have two concerns:
    1. How long will that be? A lot of the services and software (e.g. analytics) are clearly overhyped.
    2. Camera improvements (e.g., lower cost IP cameras, the move to H.264, increasing resolution at lower costs) have been and will continue to be a critical force in improving video surveillance.
    Let’s take your 3 examples:
    – The homeowner: This is simply watching a camera remotely (a functionality that has been available for 10 years); bandwidth has not changed much but the costs and ease of setup of cameras has.
    – Car dealerships: To me, one of the biggest developments for this segment are megapixel cameras. No fancy services, simply a multi-megapixel camera that provides high definition images that replaces the 4 to 8 cameras they would have previously needed.
    – Casino High Rollers: Idenitfying VIPs automatically is an incredibly hard technical problem (lighting variations, covering all the entrances, getting the right angles needed, dealing with hundreds of thousands of faces, etc). Maybe in limited scenarios with significant human review, this could be helpful. Nonetheless, I find this more hype than production ready technology. What is clearly being deployed successfully are megapixel cameras being reviewed in the same way operators have been reviewing video for years.
    While I am bullish in the long run that software and services will be the key driver. However, for the next few years, I am focusing on the ability of better, cheaper cameras to improve security operations.

  2. ITphysecGUY
    August 16, 2008 at 11:46 am

    Great post Steve!!! It’s about time we start discussing hardware-vs-software cost-to-value rational. The IT world has proven the model of “more devices are sold as their prices get cheaper!” The trick is making them cheaper and cheaper, while maintaining their profit margins and increasing the value they offer; ala what IT now leverages and calls web-services.
    Looking at the lessons learned by IT in the desktop switching, PC, PDA, telephony, cell phone, and even wireless networking markets… year to year, all of them have got cheaper per device or per port while leveraging a hosted or web services model. This model allows lite-weight edge devices (PCs with browsers, cell phones with text messaging & browsers, etc…) to get more and more functionality from network services (often running as apps on servers). This is currently true for surveillance recording, monitoring, switching, archiving, and distribution; all offered now as video management software running on IT’s favorite COTS servers.
    The question is… why do other more powerful & more capable devices (PCs, smart phones, IP phone sets, etc…) cost less and offer more functionality than today’s IP cameras? It seems like we’re seeing Moore’s law used to repeat the failures of the fat-app desktop, but this time for IP cameras. One can’t help but wonder if this is why Microsoft is considering a $40B+ acquisition of Yahoo, and why Google’s become so powerful! Most PC users rely on internal and Internet-based websites, IP phones rely on a virtual PBX, smart phones are pretty useless without the network, enterprise wireless access points rely on a wireless controller, and cameras rely on NVRs and software apps. So I believe you’re spot on… it’s become all about the software and services we use & need, not really the hardware we use to experience them.
    Why not use Moore’s law to drive down the cost of the camera, allow it to tap into endless network services, and offer higher quality video services to a wider array of users for a wider array of applications? Isn’t this in nearly everyone’s best interest because more cameras and recording systems will be sold to more users for more overall value?

  3. August 18, 2008 at 6:12 am

    Very nice post, per usual, but without good cameras (including the lighting and lenses, e.g. optics), good set up and a good understanding of the threats/scenes trying to be addressed the software and associated services will not reach their potential.
    20+ years ago (not a mistype) the industrial machine vision industry evolved along the same lines as the analytics market is doing now. This was followed by the same developments in the transportation industry (license plate reading, traffic flow monitoring, traffic incident detection). Interestingly in both cases other sensors (e.g. microwave, laser scanning, etc.) also came to the fore to complement video. Not surprising past is pretext.
    I still think the biggest challenge lies with the sale of the SYSTEM and to make sure its solving a customer’s problem and as a result the value comes from those organizations that can sell, design, integrate and maintain.

  4. August 18, 2008 at 12:31 pm

    I thing this is a great point to make, Steve. Of course cameras still come in a range of quality and capabilities, it is the software that really makes them shine. The recording time, picture quality, and special capabilities offered are determined by what system the cameras are on. It’s time the software systems get their due credit.

  5. Alon Moritz
    August 19, 2008 at 3:57 am

    I agree. The world of security is and has started to go more and more towards the old IT model of “thick server and thin client”. Software is the key and the package that the vendors will be selling is even more critical. One of the other posts mentioned easy implementation, that will be key yet somehow keeping the channels involved in the sale too is important. The real test is how we as security experts move our clients to the world of converged security, that is what this is all about and should be our focus. Aligning IT and Physical security is what we need to focus on.
    Alon Moritz

  6. August 19, 2008 at 6:15 am

    “Very nice post, per usual, but without good cameras (including the lighting and lenses, e.g. optics), good set up and a good understanding of the threats/scenes trying to be addressed the software and associated services will not reach their potential.”
    I really couldn’t have put that any better myself, Salvatore 🙂
    It’s a very interesting topic Steve, but in the great scheme of things, without decent images to play with, the rest simply won’t deliver to their full potential.
    There is a danger that working in a very technical environment, we can sometimes lose sight of the objectives, and in some cases the objectives themselves may not be all that glaringly obvious.
    As an example, using video analytics as a Site Management tool, perhaps to flag a bag being left in a given area, may work very well for it’s intended purpose, but it may not fulfil the archaic demands within a criminal justice system.
    Likewise, motion triggered recording may provide a convenient technical fix to a number of operational issues, but without the missing bits in between, it’s quite possible for the recordings to be challenged when presented as evidence in court.
    I’m always mindful that some clients often have little or no interest / understanding of the technology available, and yet they are interested in seeing pictures, and hopefully without breaking the bank.
    The fun part from my perspective is the pictures that they are seeing on screen, and what I’m actually seeing, can be two completely different visions.
    Being a simple soul at heart, I’m very keen to see the basics come good, so that any emerging developments in technology can easily be leveraged to their maximum potential. Until we see a situation where system designs are optimised and deployed in an intuitive way, there’s going to be an ever growing reliance on technology to dig us out of holes, that we were IMHO (as an industry) quite capable of filling in many years ago.

  1. No trackbacks yet.

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s

%d bloggers like this: