The Truth about Intelligent Video

How to separate intelligent video systems from those that are not.

In the growing field of video analytics, it can be difficult to discern between the countless security devices and software and solutions that claim to be "intelligent" and those that really are. But there are ways to separate fact and fiction when it comes to video analytics.

By Edward Troha

TO BE TRULY INTELLIGENT

If a system, device or solution is truly intelligent, 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 truly intelligent vision-based systems.

Additionally, the software must be capable of accurately separating foreground from background objects.  It must create a stream of metadata that reflects a 24/7 analysis of the view.  This is important because a truly intelligent system is one that can interpret and report on actual events, rather than just react to movement.

Finally, for a system to be truly intelligent it must be able to provide the user with relevant real-time information about events.  This is important because the human user must be able to easily and intelligently interact with the system in order to make the critical decisions that can mean the difference between proactively catching the bad guy or not.

IT’S HERE

Even with everything a truly intelligent system can do, the idea of cameras using artificial intelligence-based technology that can “watch” and interpret security video still sounds to some more like science fiction than reality.  But the fact is video analytics is a growing part of today’s security market.

EDGE-BASED VIDEO ANALYTICS

There is a proliferation of “edge-based” video analytics solutions on the market today.  Video analytics on the edge of the network, that is, in cameras and/or encoders, can deliver excellent return on investment, especially when those analytics are fully functional.  In addition to bandwidth throttling, they can enable an easier transition from analog to IP, allowing new deployments to be efficiently retrofitted to legacy systems.

THE BENEFIT

But analytics can go deeper than just the source of the video, offering a number of potential benefits when integrated in other network components.

Intelligent video in routers, that is, inside the network itself, can provide an entirely different set of value propositions.  For instance, if you can intelligently route the information based on the profile of the information itself, you are effectively leveraging analytics inside the network.

Intelligent capabilities in storage solutions present an even greater variety of benefits.  With these solutions, businesses can store only that which they really care about, or are required to store in the case of government or regulatory agencies.  Modulating storage saves hard drive space, and generally makes storage more efficient.

Analytics at the back-end can greatly improve investigative capabilities as well.  Video analytics can tag video with metadata enabling much quicker forensic analysis and video search capabilities.  In addition, a centralized video analytics capability provides the ability to switch functions on the fly, or as increased security concerns warrant.  This can be done without reconfiguration or without changing out functions on each and every affected camera.

Given these benefits, a wider variety of analytics-enabled products beyond cameras and encoders are becoming available and changing the way customers leverage the benefits of intelligent devices. 

NOT QUITE INTELLIGENT...

But buyers should beware -- There are several technologies on the market being dubbed intelligent even though they are not.  Here are two of them -- one old, one new.

The 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 what they’ve already got 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 video analytics are.  And at the same time these technologies are being touted as intelligent, they are rarely thin, meaning users can expect to make a relatively heavy investment in back-end servers and enterprise software to be able to run them.

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 come out 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 a couple of birds fly through the scene for you to turn it off for all the false alarms?

Users need to beware of “intelligent edge devices” that consist of VMD-enabled cameras and filtering software on the back-end.  The filters must be configured for object size and a variety of other parameters in order to reduce false alarms to a tolerable level.  And, despite the filtering software, the VMD on the camera is still merely reacting to movement.  In fact, calling VMD solutions “intelligent” is a contradiction in terms that only adds to the confusion that reigns in this market.

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 on them.

THE SMART ONES

The smart manufacturers, like Verint, Synectics and HIKVISION, are the ones teaming up with the technology providers who have been doing it for quite awhile, like ObjectVideo.  These manufacturers know that they’ll never be able to support the huge amounts of R&D time and money that it takes to bring a commercially-viable video analytic capability to market.  Really big names like GE, Siemens, Honeywell and IBM have either acquired smaller analytics companies and consequently their intellectual property or just given up.

Edward Troha is Director of Global Marketing for ObjectVideo (http://www.objectvideo.com/).

     Intruders in Trouble