printer friendly version

Going through the motions: looking at the history of video motion detection (VMD)

Much has been written about the potential for VMD. Initially, the enthusiasm amongst installers and end-users alike was matched only by the enthusiasm of manufacturers that seemed to suddenly fill the marketplace in an attempt to seize the 'opportunity' and claim the VMD 'space'. But then it all started to go wrong.

End-users began to experience unacceptable levels of false alarms. The time to install VMD systems - and particularly the fine-tuning and commissioning - took too long, and there was an overall impression that the cost simply was not worth the grief that went with it.

Concerns were expressed. The manufacturers thought the installers were at fault; the installers the manufacturers. In reality, both had a point. Some of the earliest technologies were over-complicated to install, but some installers were also using the technology inappropriately, in the wrong applications. Nuisance alarms from flying birds, or at night from moths flying too close to the cameras were a constant headache. Even cameras mounted on long poles, swaying in the wind would trigger an alarm, but was that the fault of the technology, or was it the fault of poor design and installation? The jury is out.

Given the history, why now should an installer look at the issue of video motion detection?

From a technical perspective, most video motion detection technologies are considerably more advanced than they were a decade ago:

* Smarter algorithms make them more capable of filtering out such things as car headlights, tree shadows, snow, rain, even the reflection of water from puddles, all of which could have previously triggered a false alarm.

* Target sizing is similarly more accurate: cats, birds, rabbits and other small mammals can be easily discriminated and ignored, and alarms generated by camera shake and scene wobbles can similarly be rejected.

Advances in technology have also helped in terms of ease of installation:

* Quick set-up features means a smaller number of parameters to be taken into consideration.

* Connectivity with other devices is also much easier. A device to another device from the same manufacturer is a straightforward Ethernet connection; a device to third-party technologies is via contact outputs and a series of interfaces.

Ease of installation has meant that the cost of install has dramatically fallen, because the speed of set up and commissioning is that much faster. The benefit of a VMD system over, say, a PIR, is that the VMD can be (largely) set up inside. Yes, a walk test is required, but if this is recorded and then replayed from the system for each camera, then this can be done without having to battle the elements come wind, rain or shine!

Unit costs have also fallen, again as a result of the march of technology. The approximate cost per unit today is the same as the cost per unit 15 years ago, so it is commensurately (ie when the cost of inflation is factored in) considerably less. The installer is still able to command a higher sale value, however, since the overall site installation costs are similar to using multiple PIRs.

One of the real advantages of video motion detection today is the help it gives in delivering a BS8418 compliant installation. BS8418 requires that detection must be in the view of the camera; since VMD only detects from the view of the camera, then compliance is by definition automatic. Not so for PIRs.

Of course, there is still an environment in which video motion detection is ideally suited: the space between two fences on a perimeter, for example, protecting a prison or military installation, where there is little in the way of undergrowth, or movement from passers-by or vehicles. But VMD is also appropriate in more everyday scenarios. VMD is about 'volumetric' detection: whatever a camera sees, it will detect. With technology such as it is, trees and other incumbents can be easily masked out if they are in the field of view, to prevent false alarming. VMD can also detect the direction of movement of a target, eg people leaving a building can be ignored, if the requirement is to know who is approaching.

VMD technologies are now more reliable, and have distinct advantages over other detection technologies in particular applications:

* Fence vibration systems do not detect vehicles entering a site.

* PIR detectors are similarly unreliable against vehicles when the heat source of the individuals inside is being masked by the lack of heat from the vehicle they are in.

* PIRs have to be seen; they cannot be disguised whereas VMD is discreet.

Although more sophisticated, in many respects VMDs are more straightforward to comprehend than before, and certainly easier to install. And even if there is an issue, help is never far away. Perhaps now is the time to take another look at VMD with an open mind.

Share this article:

Further reading: