Surveillance systems typically include advanced alarm handling options, but the full potential of these is not always realised. Whilst this functionality can be employed in a simple switch-like configuration, taking a fresh look at the possibilities can create enhanced performance.
It has often been said that video surveillance is a passive tool, only offering some level of purpose after an event has occurred. It is claimed that unless you want to simply review an incident or event, then the technology doesn’t offer much in the way of practical assistance. Typically, such opinions are offered by those who either don’t understand the technology and its application, or by those with a vested interest in offering another service or system. In the first case, it must be accepted that the ill-informed often like to appear knowledgeable. In the latter case, anyone who has to resort to a ‘negative sell’ probably hasn’t got anything of great merit to offer.
Video surveillance offers a wide variety of benefits, and it can be a passive tool, if that is what you require from it. However, it can also be a proactive tool too, if the potential on offer is fully exploited. The systems available today allow video surveillance technology to offer a proactive edge with regard to the protection of people and property, whilst additionally delivering tangible benefits for everyday business management.
Traditionally, when considering alarm handling, people tend think along the lines of a reactive approach. The terminology doesn’t help, almost insinuating that an alarm – always a negative connotation implying that something bad has occurred – then triggers some form of response after the fact. Given such perceptions, it’s easy to understand why many are less than enthralled with the process. It smacks of bolting the proverbial stable door after the horse has bolted.
Alarm handling is a wide and varied discipline, and covers a number of technologies. It can range from a basic reaction to some form of signal, through to the implementation of multi-layered actions as a result of predefined criteria being reported by an intelligent system.
Advances in technology and developments relating to interoperability have changed the scope of alarm handling to a significant degree. Indeed, when considering today’s solutions, it is best to consider the possibilities with a ’cause and effect’ mentality. This thinking allows more flexibility. Cause and effect simply dictates that an event (the cause) creates a second event (the effect). The difference is significant between this approach and that of traditional security thinking, where event handling was very much an ‘alarm/reaction’ two stage process.
If the negative connotations associated with alarm handling are removed, then it becomes much simpler to grasp how the various options can be used to create a robust and reliable solution; one that delivers true benefits, both in terms of protection and site management.
The cause
In order to start any sort of event-based action, a system must be able to determine the initial event: the cause. Modern surveillance system can achieve this in a number of ways. Probably the simplest is via basic input triggers. These can take the form of an attached hardware device, such as a detector, a relay or some other automated element.
The activity of such triggers is signalled to the system via a physical input. The most common devices used for this are motion detectors. However, any device that can effectively change state can create the signal.
Detection devices might sense motion in a protected area, either by analysing the space for signs of human activity, or by monitoring the area between two points, such as a portal or perimeter, to detect anyone entering a protected zone. The choice of detection devices is important, and must be specified with the goals of the solution in mind. However, it is worth remembering that while motion detectors might be the most commonly used option, any type of detector – smoke, temperature, moisture, contacts and even deliberately operated devices such as panic buttons, call points, intercoms or door bells – can effectively be connected to an alarm input.
For example, if an automated barrier is used for vehicular access to a site, that barrier being opened could also send a trigger signal to a surveillance system. Lighting being switched could also do the same, as could any range of automated processes. A door being opened, a freezer’s temperature rising above a certain threshold, a certain noise volume level being reached, or even no motion being sensed, can all be used.
It’s important not to become bogged down with the thinking that an ‘alarm input’ is only for alarms, in the traditional sense of the word. By linking other types of event, such as those generated by management-based systems, the solution can add significant benefits when a site is manned and active. The inputs don’t have to be security-related, although the security elements should not be compromised by such an approach.
Increasingly, video solutions can include an interface to data generating devices. These can include POS devices in retail applications, ATMs in financial institutions and logging systems in warehouses and logistics operations. As with trigger inputs, these data streams can not only be recorded along with the video, but they can also be used as event triggers. For example, POS units could store sales transaction data along with associated video, but could also instigate events if, for example, a No Sale till opening occurs.
Whilst alarm inputs and data streams are typically generated from external items linked to the surveillance solution, the video captured by the system itself can also be used to generate events. This can take the form of video motion detection, or video analytics.
Video-based event triggering is increasingly popular in a wide range of sites, and for good reason. The data that is captured contains much more information than that generated from a simple detection device. Consider a motion detector installed outside of a building. If someone approaches, they will be detected, and a signal is then sent to the controlling element of the video surveillance system. The ’cause’ is established as motion in a designated area, and the ‘effect’ can be any of a number of actions. Typically, this will be to trigger recording, or to raise the resolution of the video being captured. It might also be to alert an operator, or to initiate an alarm. Of course, it could equally be a site management task such as switching on lighting, powering up a digital information sign, opening a barrier, etc..
Whilst this is all well and good, the detector simply informs the system that motion has been detected. There is no more information. Somewhere, in the field of view of the sensor, movement has taken place. That’s it. If the detector is reset before the target is out of view, another instance of motion is generated. Is it the same source? Did it occur in the same place? What caused the event? It is necessary for the user to investigate and assess the footage of all events to fully understand any scenario.
Video-based detection can offer a higher degree of information. It can identify where in the field of view that motion was first detected. It can track individual targets, and can generate a ‘trail’ that clearly illustrates the path taken by the target whilst travelling in the protected area. It can identify where the target stops, and for how long. With much more information available, the criteria for events can be greatly expanded.
Interestingly for the user, this detailed information can also be collated into a report, and used for non-security purposes such as site management, without any need to review the video footage.
It has to be accepted that despite the use of terminology such as ‘intelligent’ and ‘smart’ when describing such systems, nuisance activations will occur. Obviously, with video-based detection, much depends upon the quality of the system and the suitability of the site. There are applications where it would be foolhardy to opt for such an approach.
If the system is essentially looking for any motion, then any motion will trigger an event! The closer you define what it is you want to be informed of, the more you can filter out nuisance activations. However, as the criteria become increasingly defined, so the likelihood of missing unexpected events rises. A balance needs to be struck between defining exact criteria for activation, and allowing everything and anything to be interpreted as a ’cause’.
A final point is that video-based detection works best with a sterile zone – an area in which activity or motion is not expected. This eliminates a high degree of everyday movement, and supposes that should activity take place, then it is an exception that needs to be notified. A sterile zone can be created by a fence or barrier, a No Entry or Private sign, or any demarcation that makes it obvious to those in the area that they should not be there. By entering the sterile zone, they are aware that a breach is occurring.
From a purely security-based stance, a sterile zone makes sense. However, if a user also wishes to use the surveillance system for business-related tasks, the thinking behind a sterile zone needs to be adapted.
Take a retail environment as an example. If video surveillance is used, then the system will be operable around the clock. Captured video can be used for a wide range of applications: protecting the business against customer and staff theft, protecting staff from real or threatened violence, monitoring health and safety issues, etc.. During opening hours, video analytics can also be used to monitor footfall through the site, tracking where customers are, identifying busy parts of the site at certain times, seeing where customers stop and which displays are popular.
A report can then be generated which allows management to understand which departments have most traffic and when. This information could be used to manage staffing levels, to relocate displays for higher impact on sales and to ensure that the store layout reflects customer trends. Analytics can also be used to manage the site in real time. For example, the data could create an alert if a queue forms, or if a cashpoint is unmanned, if aisles or walkways are blocked, etc..
When the site is closed, the building becomes – in effect – a sterile zone! The analytics then would be scheduled to create an event if human activity or motion is detected. The security aspects are preserved, but the ’cause and effect’ design means that when normal trade is taking place, the solution is still delivering benefits, albeit of a different type, to the user.
The effect
The most important thing to bear in mind when considering the second event, the ‘effect’, is actually the ’cause’. More specifically, it is a case of understanding the circumstances of the ’cause’, because depending upon those, there could be multiple ‘effects’. There might be a need to employ scheduling, so that events are treated differently during site open hours and closed hours. The user might want different actions to be initiated dependent upon staffing levels, whether the motion is a person walking or travelling in a vehicle, if the vehicle is registered as a staff vehicle or is unknown, or even if it is a car or a commercial vehicle.
Every site will have different requirements, and whilst these needs are virtually impossible to cover in one article, the fact is that there will be a surveillance solution that can handle the necessary events in some way. The flexibility available is growing all of the time, and the pace of development constantly adds new opportunities.
The ‘effect’ of an event could be automated response by a device or element of the system. Even the most basic surveillance devices usually include relay outputs, and these can allow triggering of external systems such as lights, barriers, door controls, signage, message annunciators or building management elements such as fans, power or even lift control. Essentially, if an automated device can be used, then the trigger for this can come from the surveillance system rather than from a discrete system.
The process is similar whether the ’cause’ was generated via a physical input or from the video data or associated metadata.
There are also options for the ‘effect’ to employ some form of communications. The surveillance system can upload associated data via an FTP connection, email a defined set of authorised persons, send information via an SMS message, contact a control room or remote monitoring operation, or stream video and other data to mobile devices.
Increasingly the scope of these communications is becoming global, and as the capabilities of tablets, smartphones and networks increase, so do the possibilities of how far reaching and complex the generated ‘effect’ can be.
The more you consider the possibilities, the more that terminology such as ‘alarm handling’ underplays the potential on offer. Whilst the security aspects of surveillance should never be underplayed, there is an increasing depth of functionality that exists outside of traditional security principles. Indeed, whilst the security aspects are required for exceptional circumstances, the management benefits can deliver value that is experienced on a daily basis.
Bringing it together
In recent months, Benchmark has taken a closer look at the options available from video management solutions, hardware-based recorders, communications devices and mainstream surveillance elements such as codecs and cameras. The vast majority of credible surveillance options include features and functions that allow some degree of ’cause and effect’ event management.
The surveillance sector features many entry-level or budget devices. Often, these have a degree of functionality that is ideal for small and simple applications. These will typically include alarm handling, but it is usually restricted to the alarm/reaction type relationships. This isn’t a negative point; it’s all about suitability for a given application. However, to enjoy the higher degree of flexibility that delivers greater value, the initial capital investment will obviously be higher. That said, the enhanced flexibility could allow budget to be drawn from other departments that will enjoy the additional benefits on offer.
Benchmark was able to create the most flexible ’cause and effect’ based event handling solutions using video management software, or higher end network video recorders. These allowed a combination of video-based and hard-triggered events, as well as more flexibility with regard to the secondary event – the effect – which included both security and management benefits.
When some of the scenarios were shown to end users, their reactions were positive, with many surprised at the depth of flexibility on offer. All admitted that the word ‘alarm’ had possibly limited their thinking on what could be achieved, and stated they would consider advanced event handling in the future.
Verdict
Taking a ’cause and effect’ approach to alarm handling might seem to only be subtly different to the more traditional ‘alarm/reaction’ approach, and it is. Much of the process is exactly the same, and the elements of systems used are also the same. However, what differs is the application.
It must accepted that the difference is minor, but by applying that difference, alarm handling can be realised as a significant tool that takes the surveillance solution’s benefits way beyond the confines of security.