By making use of various image processing algorithms, such as those in an effective video analytics system within a surveillance system, a variety of incidents and events can be detected in real time, boosting physical security and operational efficiency.
Video surveillance systems are typically installed to record video footage of areas of interest within a facility, around its perimeter or in outdoor areas which require observation, with a view to ‘catching’ and recording events related to security, safety, loss prevention, operational efficiency and business intelligence. Some examples could include, entry by unauthorised people or vehicles into restricted areas; tailgating of vehicles through secure checkpoints; unattended objects and removal of assets can also be detected.
Why video analytics?
But surveillance is one thing, how do we take it to the next level and turn the events we record into useful data we can act on? By defining the set of events that the surveillance system operator wants to be alerted to, video analytics software continuously analyses video in real time and provides an immediate alert upon detection of a relevant event.
Video surveillance has historically been dependent upon human intervention and control, with the system operator often providing the interface between alarm activations and the required or subsequent actions. The inability of the operator to monitor and process multiple alarm events is often cited as the weak point in traditional surveillance systems. The more recent introduction of intelligent video and alarm analysis has provided a reduction in operator workflow by the addition of pre-configured alarms, linked to pre-programmed actions, saving on manpower costs.
Who is providing capabilities?
Lets take an example. Sony DEPA Advanced Analytics currently uses innovative design approaches when featuring edge based analytics. The Distributed Enhanced Processing Architecture (DEPA) from Sony is an embedded technology providing pre-processing of image data within its camera technology and post-processing of image data within its recording and monitoring solutions. This technology will often form the premise of, or enable the execution of, video analytics.
AgentVI is another example of a software company that allows its analytic engine to run at the edge of the network, is completely interoperable and will easily integrate with, and run on cameras from other manufacturers. Agent Vi offers open architecture products that enable end-users to employ best-of-breed solutions, allowing them to implement an overall system design that can be specifically tailored to match individual needs.
Additionally, Geutebrueck’s G-Tect system, is an example of a VMS manufacturer that has developed analytic systems from the late ‘70s, and its products today feature edge-based analytics in its own cameras. This technology can however be used by other ‘simple’ cameras by allowing the analytic engine to run centrally and analysing the streams from the field in real time.
Changing the perception
It is important to note that analytic technologies form an essential core for any enterprise based CCTV system. However, video analytics is receiving a bad name. Many analytic disciplines are based on the simple assumption that there has been a change in the video image from the camera and this event might be of significance. Where the actual intelligence sits is in determining if such changes are significant or if it’s not weather conditions or another insignificant object blocking the camera view or causing a shift to the camera position.
The differentiator here is the ability of the analytic engine to filter relevant events or objects from non-relevant events. Once you’ve fixed the above challenge, technologies like video motion detection, direction detection, object size detection, left object detection and facial detection can be implemented. While they all use the above premise as a foundation, these disciplines are monumentally differ-ent and some are easier to implement than others; or more practical in a certain envir-onment than others.
It is with this that the effectiveness of the surveillance system lies with analytic engines, which filter out the ‘wants’ and ‘want-nots’ and either make or break a surveillance solution, impacting the effectiveness of the solution. For this reason, effective implementation cannot be based on a one-size-fits-all solution. There are real challenges in implementing such a solution when the installed environment does not lend itself to it.
However, while one site is not the same as another, there are certain fundamentals that can be considered when implementing such a solution. Installing a surveillance system is not just fire-and-forget, it’s more organic in nature and the system needs to grow with the environment.
Business security cannot be compromised, so if considering video analytics make sure that the analytic engine ticks all of the boxes, takes the environment into account and the solution is implemented according to your specific needs.
|Tel:||+27 11 848 9000|
|Fax:||+27 11 848 9003|
|Articles:||More information and articles about Westcon-Comstor Sub-Saharan Africa|
© Technews Publishing (Pty) Ltd | All Rights Reserved