Home > Today's Headline News

By Sunny Kim

WHAT IS VIDEO ANALYTICS?

Video analytics, or video content analysis software, comes in a variety of forms.  Video Motion Detection (VMD), facial recognition software, automated number plate recognition software and vision-based process inspection programs are all forms of video analytics.  In this group also is intelligent video analytics, or intelligent video.

Put simply, intelligent video analytics is software that extracts useful information from video imagery by turning video into data.  The underlying science, called computer vision, dates back to artificial intelligence research from the 1960’s.  It has born fruit only after decades of research and funding by such forward-looking organizations as the U.S.A. Defense Advanced Research Projects Agency, which was also responsible for funding the development of the Internet.  The computer vision discipline includes sophisticated approaches to modeling the world as seen through the lens of a video camera.  Environmental phenomena such as wind, rain, sunshine and shadows are detected, analyzed and rejected as immaterial natural occurrences.  Objects of interest such as people and vehicles are automatically identified and tracked through the camera’s view.

Intelligent video software uses these computer vision techniques to detect and track objects, and then analyzes them to identify specific behaviors.  For example, intelligent video can detect when a person is breaching a perimeter or loitering around an ATM, or when a car is parked in an illegal space or for an unusual length of time.

There are three important criteria that will enable a determination if a system, device or solution is truly intelligent:

First, it must be an application of artificial intelligence, specifically, computer vision.  This is important because computer vision, the science of teaching a computer to see, is the basis for all truly intelligent vision-based systems.

Second, the software must be capable of accurately separating foreground from background objects reliably.  This is important because a truly intelligent system is one that can interpret and report on actual events, rather than just react to movement.

Third, the system must be able to provide the user with relevant information about these events, ultimately in the form of meta data -- a complete description of everything that is happening within the scene at every moment.  This is important because this data provides the user with the basis for highly efficient forensic searches and the ability to leverage the video infrastructure to obtain highly valuable business intelligence.

The ability to intelligently automate the process of monitoring video feeds may sound like science fiction, and considering the claims made by some vendors, it is possible that unrealistic expectations about the current capabilities of the technology have surfaced.  What can you do?

When confronted with an unlikely claim about video analytics, a good rule of thumb is to ask the following: “Could a human watching the video perform the same task?”  If the answer is ’no’, then it is unlikely that any form of video analytics can do it either.   Even if the answer is ’yes’ it is still possible that the abilities of video analytics will be tested to the maximum.  Remember, human vision has been evolving for about five million years.   Computers have only been doing this for about forty!

But even with the limitations of intelligent video, it is today being used in a variety of sectors, including airports, seaports, borders, process manufacturing, facilities management, retail and banking.  It is helping to increase security effectiveness in a smarter, faster and more efficient way.  For instance, intelligent video is used in the rail transport sector to secure rail yard perimeters, monitor platform crowding, reduce vandalism, ensure passenger safety with in-car cameras, prevent gate-jumping and address wrong-way ingress and egress, among other applications.

THE OLD, THE NEW AND THE WISE

So, what are examples of technologies being touted today as intelligent?  Here are two of them:

A newer one is the biometric technology facial recognition, which has a close relative in Automated Number Plate Recognition (ANPR).  Although both of these are vision-based, and do analyze what they are seeing, they are probability-based.  That is, they rely on the relative quality of the view to do their jobs and then can only compare what they have with images they already have in the database.  So unless you think a relational database application is ‘intelligent’  facial recognition and ANPR are not intelligent in the way computer vision-based intelligent video is.  And at the same time, these technologies are being touted as intelligent, they are rarely thin -- expect to make a heavy investment in back-end servers and enterprise software to use facial recognition or ANPR in your video surveillance environment.

The other technology that even more frequently is pitched as intelligent is Video Motion Detection, or VMD.  New flavors and permutations of this dinosaur seem to be released all the time, but don’t be fooled by VMD -- it’s definitely not intelligent.  Why not?

VMD does not analyze; it reacts to motion.  It does not interpret events; it reacts to motion.  It does not classify objects; it reacts to motion.  So VMD is aptly named -- in its basic form, any movement of any pixel at any time will cause an alarm.  How many times must the wind blow tree branches within the view, or birds fly through the scene for it to be shut down for all the false alarms it generates?  Interestingly, one major camera manufacturer puts VMD in the camera -- claiming it to be an ‘intelligent edge device’ -- then requires installation of proprietary filtering software on back end servers.  The filters must be painstakingly configured for object size and a variety of other parameters, all in the cause of reducing false alarms to a tolerable level.  Despite all that the filtering software does, the VMD on the camera is still merely reacting to movement -- definitely not intelligent.

A similar approach has even been announced by a major EU-based security company which manufactures lots of devices and systems, claiming that the total system is the only way to achieve ‘intelligence’.  What manner of ‘intelligence’ do the really have?   

They even name their solution ‘Intelligent Video Motion Detection’ -- a contradiction in terms that only adds to the confusion that reigns in this market.

MONEY TALKS

Software companies, device manufacturers, end-to-end solution providers -- they’re all claiming intelligence for one reason: money.

Slapping VMD into a device that previously didn’t have it, then calling it ‘intelligent’ and adding as much as a 30% price premium won’t get the end-user the truly intelligent capabilities they require.  These companies are late to the game and are frantically trying to catch up.  They attempt to do so by creating noise and confusion in the market, rather than by creating products that have real video analytics embedded in them.

To address this, manufacturers like Tyco International, Verint and March Networks and others are teaming up with the intelligent video technology providers who have been doing it for quite awhile, like ObjectVideo, because these manufacturers know they’ll never be able to support the huge amounts of R&D time and money that it takes to bring a commercially viable intelligent capability to market.  Big names like GE, Siemens, Honeywell and IBM have either acquired smaller analytics companies and consequently their intellectual property or just given up and put their money on the VMD they already have.

WHY BUY IT?

First, a prospective buyer should ask the tough questions: Are the analytics computer vision-based or are they video motion detection?  Are the analytics capabilities full-featured or just a few algorithms that have limited applicability outside the specific scenarios for which they were developed?  What are the vendor’s abilities to provide training, configuration and full technical support to help ensure the user gets the most from its analytics investment?  In the security model, the real value proposition is surprisingly more complicated than just catching the bad guy.  Early adopters have been willing to try new technologies, but more conservative buyers require a measurable return on investment before they are willing to upgrade their systems.  Sophisticated buyers are creating video analytics strategies that define the role analytics will play within the organization.

A strategy starts with understanding how analytics will affect your mission -- what can it do for you?  Obviously efficient data gathering for security or business intelligence is a reason for adopting truly intelligent analytics.  But there are other fundamental drivers.   Video analytics can have a significant impact on the efficiency of your personnel and your infrastructure.  If your people are not assigned to monitoring stations, they can be more mobile.  This means they are more visible to act as a deterrent and can provide faster response should an incident occur.  Furthermore, video-based investigations can be automated with intelligent video so that they take minutes rather than hours.   Understanding which analytics features and functions you require is a key step to selecting appropriate analytics vendors and products. A second part of a strategy involves integrating video analytics with existing CCTV infrastructure to increase the overall system value.  Over recent years, businesses have begun to switch their CCTV infrastructure from legacy analogue video to new IP-based digital video systems.  This means that video data, which was once completely ignored on coaxial cable networks, can now be completely ignored over IP networks.  And video can consume massive amounts of network bandwidth and digital storage in the process.

INTELLIGENCE AT THE EDGE AND THEN SOME

A lot has been written about intelligent video analytics at the edge of the network and it’s true that there is great value in putting analytics in cameras, IP-video encoders and other front edge devices.  A vast majority of video imagery is completely useless for security or business intelligence, but still consumes large amounts of network bandwidth and storage.  Putting intelligent analytics at the edge provides a mechanism to filter the video at the source.  When nothing important is happening, video can be transmitted and stored at a lower frame-rate, resolution and quality or be deleted.  Only video deemed important need be transmitted and preserved at high quality.  This is represents significant savings of bandwidth and storage.  Highly valuable.

It is also valuable to put intelligence inside the network itself -- such as within a network switch or router.  If the network infrastructure understands the content of the data flowing through it, it can make sensible decisions about what information is important, how to route it and how to optimize quality of service to the user of the network.  Intelligent analytics embedded in a router gives the network itself the ability to peer inside a major data stream -- video -- and intelligently determine the importance of the content.  Also highly valuable.

Finally, intelligent analytics also has a significant play at the back-end of the video network -- in a video storage device or management system.  The same case for bandwidth throttling applies to storage capacity, and putting intelligent analytics at the back-end allows a video management system to save only the important video information.  In addition, many customers don’t want the same analytics applications running on the same set of cameras all the time.  Having centralized intelligent analytics allows customers to configure different application sets on different cameras at different times.  And finally, putting analytics in the storage solution allows a device to tag stored video with analytics-derived meta data that can make video investigations significantly more efficient.  It would seem very prudent to attempt a look at the future when considering the role for intelligent video within a security operation.  It’s one thing to select analytics vendors and products that fit your mission today, but be aware that as you use the technology, and begin to derive benefit from it, new uses will present themselves. 

Make sure to select a partner that can grow with you as the use of intelligent video becomes a more significant part of your operational paradigm.  And for the same reason, make sure to select an analytics provider who can offer a full set of truly intelligent capabilities, not just an a few algorithms that are sold individually.

WHAT TO LOOK FOR?

Understanding there is hype surrounding video analytics, so there are some things to look out for if considering using the technology as an end-customer looking to integrate analytics into infrastructure and process.  Look for strength in three separate areas: technology, product and company.

Seek technology that is proven in the real world.  Make sure it is capable of providing the applications you require in environments relevant to your mission.  A flashy and compelling sales presentation does not guarantee strong operational performance.   

Don’t agree to be someone’s pilot project.  Look for truly intelligent analytics products.

The software must have well-tested features and functions that support your mission, and be able to seamlessly integrate with your preferred CCTV infrastructure.  You don’t want to buy hardware products from analytics vendors nor do you want home-grown analytics produced by a hardware vendor.  Make sure you choose best-of-breed intelligent analytics functionality from a market leader and a hardware platform from an equipment market leader -- and make sure they are integrated together.  Look for an analytics company that has a real track record and is capable of providing the support that’s needed.

So ask pointed questions of providers who claim intelligent products according to the three criteria.  Insist that the intelligent analytics you acquire within the solution have comprehensive functionality (not just a few analytic capabilities), are very easy to integrate into any existing infrastructure and are consistent with a video analytics strategy that supports the fulfillment of a security mission.

MAKE AN INTELLIGENT DECISION

Everyone loves video.  It’s cheap, intuitive and ubiquitous.  Unfortunately, video systems have had thousands of eyeballs trying to watch, with unacceptable results.  Intelligent video analytics gives the power of intelligence to CCTV systems by extracting useful data from video, and reporting what’s truly relevant.

Deployed properly, as a key part of a comprehensive security solution, intelligent video technology can turn eyes-everywhere cameras into valuable data sources, allowing you to make quick assessments and real-time, thoughtful decisions. 

It really is up to the end-user to drive how intelligent technology works within their video operation.   Know that intelligent video can now provide real value.  And don’t worry; the time of the pushy vendor is over.  Now is the time for the wise buyer.

Sunny Kim is Editor of SecurityWorld INT’L (www.securityworldmag.com).  Please send your comments to kbs@infothe.com.