If only the CCTV image was in focus! It would be a welcome change as, sadly, the majority of CCTV cameras are out of focus, poorly positioned, poorly installed, poorly lit and poorly utilised. Sam Samuels endeavours to provide security managers with the tools to be able to define requirements, and to audit and assess video surveillance systems (VSS) according to the level and type of risks posed in their particular environment.
Activity detection, facial recognition, licence plate recognition, high-definition camera formats… With an abundance of newly-emerging capabilities to choose from, those seeking to invest in a new CCTV system may (understandably) feel like they are faced with an impossible task in selecting the correct security solution for their particular needs.
CCTV systems for airports, as for other public spaces, tend to be procured by means of a bidding process; there is often a prequalification stage, usually based upon the financial reliability of the bidders, and the project will inevitably be awarded to the lowest bid offered by the pre-qualified companies. Thus, some systems are designed to meet budgetary requirements rather than to meet security standards. Indeed, procurement will often proceed without the customer even being made aware of basic industry standards. Additionally, the bidders may not be concentrating on what the client actually needs, but on what they are trying to achieve in terms of their own business model. Therefore, it is imperative that every security manager has an understanding of the requirements and limitations of CCTV systems prior to implementing an installation or upgrade.
This article aims to provide the aviation security manager with the knowledge, tools and, most importantly, the correct mindset to be able to assess a design or audit an installation. The most important thing to remember is that 95% of CCTV image problems are in fact lighting problems.
What is CCTV?
The most basic CCTV system comprises the following:
2. Cables and power
3. Display monitor
4. Adequate lighting
To this may be added:
5. Recording/replay and recording medium
CCTV cannot be regarded as a security measure in itself. It is merely a tool, which relies upon effective security personnel to act upon the information derived from the CCTV system. There are three uses for a CCTV system: deterrence, live monitoring (including number plate recognition) and post-incident investigation (including identification of persons). All three of these rely upon the effectiveness of the overall system design to achieve optimum performance according to a client’s operational requirements.
To act as a deterrent, the potential offender must be aware that the CCTV system exists, that there are first responders available to intervene, and that evidence of any potential infringement will be captured. Further, the value of the CCTV will vary according to the immediate goals of the offender. For example, CCTV will have no deterrent value for a suicide bomber; however, for a prospective suicide bomber conducting hostile surveillance, CCTV may be effective.
An actively-used CCTV system can be used to detect hostile surveillance and generate a security response. Effective live monitoring relies upon vigilant, motivated and well-trained staff. Even so, there is no guarantee that an incident will be detected and responded to correctly. In an experiment conducted by [insert name here] at [location], CCTV operators were [instructed to] identify a target carrying a red umbrella. Operators viewing one, four, six and nine monitors achieved differing accuracy detection scores:
|Number of screens monitored||Accuracy|
Video analytic systems with movement and missing object detection capabilities can ease the task of monitoring staff but nothing can guarantee detection of an incident.
The most important use of CCTV is in post-incident investigation. In recent terrorist incidents in London, the police determined that the fully functional pan-tilt-zoom (PTZ) cameras provided no useful evidence, and all of the evidential images were subsequently derived from static cameras at transport hubs (stations, etc.) and shops and businesses overlooking public pavements/doorways.
I have seen airport surveillance schemes relying solely upon a multitude of constantly-moving PTZ cameras. The panning images leave monitoring staff unable to detect any sudden movement, which might be indicative of an incident, since everything within the image is moving. Worse than useless!
Post-incident investigation relies principally on good quality, fixed cameras at strategic points providing correct, well lit, well aligned and focused images, which may be properly stored and recorded without excessive compression. Excessive compression of images – either for transmission over a network or to maximise storage – is a serious error. An excellent video stream can be reduced to a pixelated mess through thoughtless design or user error.
Standards have long existed to help ensure the appropriate use of CCTV in aviation applications but these standards are poorly understood, and the average aviation security manager does not have the tools to assess the quality of their CCTV systems.
International AVSEC applications are subject to the International Civil Aviation Organisation (ICAO) Standards and Recommended Practices (SARPs) of Annex 17 and the Aviation Security Manual (Doc. 8973), and the same principles should be applicable in domestic, cargo and general aviation (GA) contexts.
Sadly, the advice on CCTV from ICAO is less than adequate, not having been revised for the past 20 years. Apparently, ICAO still believes that CCTV is recorded on reel-to-reel tape. The advice within the Aviation Security Manual occupies only three and a half pages of an 818-page document.
Whilst the International Air Transport Association (IATA) has demonstrated itself to be far more dynamic in terms of promoting security awareness than ICAO, it offers a similar standard of advice regarding CCTV. The European Union (EU) is little better: Section 4.6 of the document, Commissioning Implementing Regulation (EU) 2015/1998 of 5 November 2015 laying down detailed measures for the implementation of the common basic standards on aviation security asks, ‘Does the building have an effective CCTV system?’ Yes or No. This is an important question, yet the document fails to define what is meant by ‘an effective CCTV system’.
A breath of fresh air was provided in 2018 when the Airports Council International (ACI) published the Landside Security Handbook developed in collaboration with Arup. However, even this excellent handbook offers little to assist security managers in actually implementing, commissioning, assessing and auditing their CCTV Systems.
Perhaps the ‘least bad’ source of CCTV guidance can be found in the respective Transportation Security Administration (TSA) and Safe Skies guidance. The TSA guidance contains perhaps the most elegant guidance for validating the positioning of CCTV cameras: “Selection of each location and camera type depends on operational mode and local topology.” Unfortunately, whilst this provides good guidance on the purpose of cameras in specific areas and applications, it does not specify the image format in assessable and auditable terms.
CCTV has been a security tool for many years, and the best tool for assessing the quality of CCTV images has been the Rotakin Test, a 1600mm x 400mm target developed by the UK’s Home Office in the 1980s. Use of the target was included in the British Standard 50132, and this then evolved into a European, and eventually a global, ISO Standard. ISO 50132 defines the quality of CCTV images according to the percentage of the screen occupied by the standard Rotakin target. However, CCTV at the time was based on black and white analogue technology and, as the security industry transitioned to colour images in digital formats, Rotakin and ISO 50132 became obsolete. Rotakin, while still available today, has been replaced by better targets, and 50132 has been replaced by IEC 62676, which states that test targets must allow the measurement of both colour and pixel density. However, while the Rotakin test is effectively now redundant, the original concept of a standard target of known size against which images could be measured remains central to modern video analysis methods.
An A4/letter-sized CCTV target that is compliant with IEC62676-4 may be downloaded from the ASI website (at XXXX), and can be used to both audit and commission CCTV systems. Users can print the target on a conventional office printer and place them throughout an environment – determining pixel density and depth of image and focus.
The operational usage and environment of the CCTV system should dictate the type of target being used to test each camera. For example, an external camera may be used to detect a vehicle rather than a person and so the field of view should be adjusted to account for this.
IEC 62676-4 – What you should know
IEC 62676 is the international standard for video surveillance systems (VSS) – note the term ‘CCTV’ is no longer in use – and may be regarded as the de-facto global standard for the design and implementation of VSS. The International Electrotechnical Commission (IEC) standards may be regarded as obligatory while the ISO Standards tend to be regarded as advisory. With few exceptions, most countries are members of both ISO and IEC organisations and the use of this standard should be regarded as an obligation, subject to local standards and regulations, which may stipulate alternative requirements. This applies to those countries aligned with US Standards which have been accepted by the American National Standards Institute (ANSI). (Visit https://webstore.ansi.org/Standards/IEC/IEC62676Ed2014 for more information).
‘IEC 62676-4 Video surveillance systems for use in security applications – Part 4: Application guidelines’ sets out six types of VSS image, which are defined primarily in terms of their application goal (to monitor, detect, observe, recognise, identify or inspect). Each application determines the ideal target screen height in pixels per metre (ppm). The target size referred to in 62676-4 is 1600 x 400mm.
To monitor or crowd-control, the target shall represent not less than 5% of screen height (both PAL and NTSC resolution), or more than 80mm per pixel = Min. 12.5ppm.
In order to detect, the target shall represent not less than 10% of screen height (both PAL and NTSC resolution), or more than 40mm per pixel = Min. 25ppm
To observe, the target shall represent 25% (PAL) or 30% (NTSC) of screen height, or more than 16mm per pixel = Min. 62.5ppm.
To recognise, the target shall represent not less than 50% (PAL) and 60% (NTSC) of screen height, or more than 8mm per pixel = Min. 125ppm.
To identify, the target shall represent not less than 100% (PAL) or 120% (NTSC) of screen height, or more than 4mm per pixel = Min. 250ppm.
To inspect, the target shall represent not less than 400% (PAL) and 450% (NTSC) of screen height, or more than 1mm per pixel = Min. 1000ppm.
The two different assessment criteria provide different field-of-view widths and, at larger chip resolutions, the disparity becomes acute and the pixel size criteria must be discounted for ‘monitor’ and ‘detect’ applications.
The ‘monitor’ and ‘detect’ image types are designed for use in perimeter alarm systems and as such, notwithstanding the performance of the individual camera, must provide the correct percentage screen height to remain effective.
The percentages provided for the first five applications are for fixed and fully functional cameras whilst the ‘inspect’ image type is used mainly for full functional cameras or special applications such as ATMs, bank teller windows and casinos.
Sam Samuels is a former police officer with extensive experience in hostile environments across the world. He has been working with commercial video surveillance systems in aviation, law enforcement, leisure and critical infrastructure applications since the early 1990s. As well as technical considerations, he focuses on the psychology of crime, particularly regarding adaptable offenders. He has developed a portfolio of CCTV Targets for easy client use, including facial recognition charts and life-sized facial images, 3D-printed facial targets, a full-sized colour replacement for the Rotakin target, a laptop-sized, fully-functional CCTV target and the downloadable aide memoire and A4/Letter sized target accompanying this article.