Video Analytics: from access control to behaviour detection and beyond

20 Dec 2011

Daniel Wan assesses the development of video analytics from a perceived premium add-on to a cost-effective and key component in certain security applications.

The use of analytics has evolved beyond fulfilling basic security needs to include advanced features such as traffic monitoring functions and detection of left luggage within transportation hubs.

On a basic level, the ability of analytics to identify the movement and speed of vehicles and people, as opposed to other spurious objects, lends the technology to monitoring of perimeters and suspicious behaviour within defined areas. Clearly, there is a requirement for security alerts where there is a danger of cars or pedestrians entering hazardous or restricted zones, such as airport runways.

Left luggage poses a specific security threat in airports and, as such, has become an increasing area of interest for airport operators and governments. In the United Kingdom, the Home Office’s Imagery Library for Intelligent Detection Systems (i-LIDS) certification contains a specific scenario relating to detection and tracking of objects. Video analytics can alert security guards to packages and luggage left within a scene, which otherwise may go undetected in the midst of multiple CCTV screens.  Similarly, the software can highlight instances of theft, where objects have been removed from a scene.

Wider Acceptance

Recently, a number of factors have contributed to an increase in interest and demand, enabling video analytics to be reconsidered as a viable video surveillance option. The growing transition from analogue to networked CCTV has seen a wider acceptance of digital solutions, including analytics, leading to further benefits in terms of being able to monitor and manage systems from remote locations.

More importantly, end users and integrators were historically unable to refer to a standard, or accreditation, for analytics systems in order to judge their effectiveness before purchase. This was the case until the introduction of the Home Office’s i-LIDS certification.

i-LIDS is the UK government’s benchmark standard for video analytics technology and is awarded to security products judged to have met the stringent criteria specified by the Home Office Scientific Development Branch (HOSDB) in the development of video based detection systems for government use.

Manufacturers meeting the highest level of performance classification receive the i-LIDS accreditation, which is the first real independent measure of the quality and functionality of a video analytics system.

Improving the Security Function

When the costs of manned guarding and operators are considered, it’s tempting when assessing analytics to assume that the number of security staff can be immediately reduced by introducing the technology. Perhaps it’s more appropriate to view video analytics as a way to improve the efficiency of operators and guards.

It has been proven that CCTV operators begin to miss significant events after monitoring video footage for more than twenty to thirty minutes. This effect is multiplied by the large number of screens that operators are now required to monitor in a control room environment.  Analytics can be used to assist in highlighting notable events, thereby allowing operators to do what they do best; deciding whether an incident is suspicious or not. Security functions can therefore handle more cameras more effectively with their existing staff.  The job of monitoring is also made more stimulating, removing the need to constantly survey a wall of video screens for long periods.

In replacing guards with yet more cameras, the net effect is to create more footage and screens for the operators to monitor.  Supporting the guards with analytics means the ability to switch guards to mobile patrols based on analytics incidents, covering a larger area more effectively and responding to real incidents more quickly, rather than just following a routine guard tour. Above all, analytics also enables security teams to highlight potential risk situations that are taking place and have patrols pre-empt and prevent incidents rather than just respond after an incident has occurred.

Perimeter Protection

There are a number of video analytics packages available in the market today, all offering a multitude of capabilities, and not always linked to security.  However, the majority of installed systems have focused on one of video analytics’ core competencies: perimeter intrusion detection systems. Security officials at airports around the world need to deliver powerful intrusion detection systems that secure their sites perimeters, provide electronic access control, and supply integrated video assessment solutions for unmatched site protection.

These can supplement fences, or replace external PIRs, seismic detectors etc., allowing the user to identify specific areas where intruders will be identified. Potentially this includes virtual ‘fence’ lines that will trigger when an intruder climbs over it (rather than when a guard patrols along it), tripwires that trigger when crossed in specific directions and alert areas, such as nearby roads, which will trigger if a car-sized object dwells in them for too long. Combining these rules ensures only suspicious behaviours trigger the alarms and not spurious objects such as the local wildlife or weather related effects.

Eliminating such false alarms in a sterile zone is a primary test in the HOSDB’s evaluation of video analytics systems. This specific scenario continues to be a much demanded i-LIDS accreditation as increasing numbers of end users and integrators realise the performance and cost benefits of video analytics in perimeter detection type applications.

Return on Investment

Although attracted by the benefits of video analytics, security managers have always faced challenges in justifying the investment in what has been perceived as a premium technology. One method of providing a clear return on investment is to look at the existing costs for a business, or its losses.

Prevention is almost always cheaper than cure. Although not a universal panacea, the ability of analytics to highlight suspicious behaviour and let operators react to and prevent them from turning into incidents rather than just reviewing video to help catch suspects or manage incidents after the event, can provide quick returns in avoiding loss and damage.

Daniel Wan is Channel Marketing Leader UK at Honeywell Security Group. Honeywell’s new Active Alert v4.7 video analytics solution, has been evaluated by the UK Home Office Scientific Development Branch (HOSDB) and awarded i-LIDS accreditation both as a primary detection system in an operational alert role and as an event based recording system for sterile zone monitoring applications.

READ FULL ARTICLE ON THE DIGITAL EDITION>>>

One Response to “Video Analytics: from access control to behaviour detection and beyond”

  1. [...] for more than 20 minutes to 30 minutes, notes Daniel Wan, a manager with Honeywell Security Group, in this article. The large number of screens that operators now must monitor in a control room environment [...]

Leave a Reply