SPF VSS Standard for Buildings: A Practical Guide to CCTV Design in Singapore

A CCTV system should not begin with the question:

“How many cameras do we need?”

The better starting point is:

“What security problem are we trying to solve, what must the camera actually show, and what needs to happen when an incident occurs?”

That is one of the most important principles behind the Video Surveillance System Standard for Buildings, prepared by the Singapore Police Force with inputs from HTX and the Ministry of Home Affairs.

The Standard is intended to support the use of Video Surveillance Systems, or VSS, to improve the safety and security management of buildings. It describes how cameras can assist with early detection, response, business-contingency coordination, investigations and evidence. At the same time, it explicitly says VSS should not be the sole protective measure and should operate together with measures such as access control, intrusion detection and security response.

The Standard’s overall philosophy can be summarised as:

Define the security problem first. Design the surveillance system around the required operational outcome. Then prove that the installed system actually achieves it.

What is the SPF VSS Standard for Buildings?

The Video Surveillance System Standard for Buildings provides recommendations for the specification, installation and operation of VSS across buildings in Singapore.

The version used for this article is Version 2.0, March 2022, prepared by the Singapore Police Force with inputs from the Home Team Science and Technology Agency and Ministry of Home Affairs.

It covers areas including:

• security lighting;
• camera requirements;
• PTZ and IR cameras;
• camera tamper protection;
• image presentation;
• operator monitoring;
• video analytics;
• recording;
• storage;
• transmission;
• power;
• integration;
• commissioning;
• maintenance;
• field of view;
• coverage requirements;
• operator training;
• signage; and
• cybersecurity.

So it is considerably broader than simply specifying a camera resolution.

Is the VSS Standard mandatory?

The document itself describes the VSS Standard as a set of recommendations intended to guide building owners and provide a consistent approach to VSS specification, installation and operation.

It should therefore not simply be described as a universal legal requirement for every CCTV installation.

Where a particular development, authority, contract or regulatory regime separately makes certain provisions compulsory, that should be established independently.

For the purposes of this article series, we will distinguish between:

what the SPF VSS Standard recommends

and

what may separately be required by legislation, regulation, contract or another applicable standard.

What types of buildings does the Standard apply to?

The Standard says it applies to VSS installations at various types of facilities and buildings, including hospitality and retail premises and government institutions.

Its principles can therefore be relevant to a wide range of building-security projects.

However, not every camera or every location necessarily needs the same surveillance objective.

The required coverage depends on what the camera is expected to achieve.

What is the purpose of CCTV under the VSS Standard?

The Standard describes the purpose of VSS as capturing, handling and displaying video information so operators have sufficient information to:

detect anomalies, support real-time operations and support post-incident investigations involving safety and security incidents.

It also describes VSS as supporting:

early detection, security response, business-contingency coordination, investigation and evidence.

This is a broader purpose than simply:

“record video in case something happens.”

Can CCTV be the only security measure?

No.

The Standard expressly says VSS does not perform an active protective role and should not be designed as the sole protection for an area.

It should operate with measures such as:

access control, intrusion detection, fence intrusion detection and security response.

A camera can show somebody forcing a door.

It cannot physically stop that person from entering.

Consultant’s perspective

A useful way to think about CCTV is:

CCTV helps you see, understand and investigate. Other security measures may be needed to physically delay, deny or respond.

What are the four stages of CCTV design under the Standard?

The Standard establishes four key stages:

1. Define the problem
   2. Define the requirements
   3. Develop the technical specifications
   4. Commission and verify the system

The final stage requires the building owner to verify that the installed system meets the operational requirements and is fit for purpose.

This is probably the single most important framework in the document.

Stage 1: What problem are we trying to solve?

The first stage is to define the underlying:

security threat, public-safety issue or other vulnerability.

The Standard even says the building owner should consider whether CCTV is actually the appropriate response, or whether another solution may be more suitable.

This means the design should not begin with:

“Put one camera in every corner.”

Instead, the designer should understand:

What event do we need to detect, monitor, verify or investigate?

Why is this different from simply doing a CCTV layout?

Because a drawing showing camera icons does not explain what each camera is supposed to achieve.

For example, a camera could be intended to:

• monitor general activity;
• identify a person entering the main entrance;
• observe a car park;
• capture a vehicle number plate;
• verify an intrusion alarm;
• monitor a Critical Asset; or
• support post-incident investigation.

Those are different operational objectives.

They may require different:

camera positions, lenses, image detail, lighting and recording settings.

Stage 2: What are Operational Requirements?

The Standard says the second stage is to define VSS requirements by understanding the areas of concern, operational issues and required responses before deciding on the system requirements.

A useful Operational Requirement might therefore be:

Provide a frontal image of people entering through the main entrance at sufficient quality for identification.

That is much more useful than:

Install one 4MP camera above main entrance.

The first describes the required security outcome.

The second merely describes equipment.

Why shouldn’t camera specification come first?

Because the equipment should be selected to satisfy the Operational Requirement.

The Standard specifically says individual components must meet the operational requirements, work together and allow the VSS as a whole to meet those requirements.

That includes more than cameras.

It can involve:

lighting, transmission, network infrastructure, displays, recording and storage.

Consultant’s perspective

This is why a CCTV specification should ideally come after the required views and outcomes have been established.

Stage 3: What goes into the technical specification?

Once the required surveillance outcome is known, the technical specification can define what the system needs in order to achieve it.

The VSS Standard covers technical areas such as:

camera resolution, field of view, lighting, WDR, camera type, recording resolution, frame rate, storage, transmission, power, system integration and maintenance.

The technical specification should therefore serve the Operational Requirement rather than become the objective itself.

Does the VSS Standard specify a minimum camera resolution?

Yes.

The Standard says cameras should have a minimum resolution of HD 1080p, 1920 × 1080 pixels or equivalent, together with true Wide Dynamic Range capability for challenging lighting conditions.

But this must not be misunderstood.

A 1080p camera meeting the minimum camera resolution does not automatically mean every person within the image can be identified.

That depends on:

field of view, distance, target size, lighting, angle, compression and other factors.

Is a higher-megapixel camera automatically better?

Not necessarily.

Resolution is only one part of image quality.

The VSS Standard specifically warns that the required image detail depends on factors beyond nominal resolution, including:

lighting, angle of view and compression.

A high-resolution camera pointed at too wide an area may still provide insufficient pixels on the person or object of interest.

This is why camera design should be based on required field of view and level of detail, not the megapixel figure alone.

What are Detect, Observe, Recognise and Identify?

The Standard uses four Field of View categories:

Detect
Determine whether a person is present.

Observe
See characteristic details such as clothing while retaining enough surrounding context to understand activity.

Recognise
Determine with a high degree of certainty whether someone is the same person previously seen.

Identify
Provide sufficient image quality and detail for the person’s identity to be established beyond reasonable doubt.

These are different surveillance outcomes.

A camera intended to monitor general activity does not necessarily need the same level of detail as one intended to identify everyone entering a building.

Does one camera need to provide Identification everywhere in its image?

No such universal requirement appears in the Standard.

Different locations are assigned different coverage expectations.

For example, the Standard says general common-area coverage should achieve an Observation level, while frontal views at main entrances and exits should achieve Identification.

That distinction is critical.

The question should not be:

“Is this an identification camera?”

but rather:

“At what location within this camera’s field of view must Identification be achieved?”

What CCTV coverage does the Standard recommend for common areas?

The Standard says general access areas such as:

main entrance lobbies, streets and pavements within the development, car parks, loading and unloading bays, and vehicle boarding or alighting points

should have sufficient cameras to provide comprehensive coverage.

It says general common-area coverage should achieve a minimum Observation level.

It also says camera positioning should account for periods of maximum human density.

This is important because a camera view that looks excellent in an empty building may perform very differently when the area is crowded.

What CCTV coverage is recommended at main entrances?

The Standard says frontal views of people entering or leaving through main entrances and exits should achieve a minimum Identification level.

It also says entrance cameras should look towards the doorway or driver rather than simply looking down from above.

That is a particularly important design point.

A camera may technically cover the entrance but still fail to capture usable facial images because of its angle.

What about emergency exits?

The Standard says there should be frontal coverage in both directions at emergency exits, with entrances to the escape routes also covered from public areas.

General views of emergency exits should achieve at least Observation level.

This reflects the fact that emergency exits can also become alternative access routes.

What about vehicles?

For vehicle entrances and exits, the Standard says cameras should capture vehicle number plates at locations such as:

car-park gantries and loading/unloading bay entrances or exits.

The technical design required to achieve reliable plate capture is a more detailed surveillance-design topic that we will cover later in this VSS series.

What about lifts, staircases and counters?

Where lifts or stairs act as alternative entry or exit points into the building, the Standard recommends frontal and general views at Observation level.

For locations involving security checks or registration, such as checkpoints or ticket counters, frontal views should achieve Recognition level.

This demonstrates how the required image detail changes according to the operational function of the area.

What about sensitive and Critical Areas?

The Standard distinguishes between sensitive rooms and Critical Areas.

For sensitive rooms, external access areas should receive Observation coverage. Where a door has an intrusion alarm, activation should automatically display the relevant camera image, with the door view achieving Identification level.

For Critical Areas such as rooftop or air-intake access points, the external access area should provide Observation coverage while door views should achieve Identification. Internal areas housing Critical Assets should achieve Observation coverage.

This is a strong example of how CCTV and intrusion detection can work together.

Why does the VSS Standard pay so much attention to lighting?

Because a camera cannot produce useful images without suitable illumination.

The Standard says security lighting should be identified early so that lighting can support the required image quality under different operating conditions.

It also warns that poorly designed lighting can create glare and actually reduce the ability of both cameras and security personnel to see properly.

Consultant’s perspective

A common mistake is trying to solve poor nighttime image quality by changing the camera when the real problem is:

the lighting environment.

Should CCTV cameras be visible?

For cameras in public areas, the Standard says they should be positioned where they cannot easily be:

evaded, damaged or obscured

and should be clearly visible to the public.

Visibility can support deterrence while the mounting position can reduce opportunities for tampering.

Should cameras meeting coverage requirements be fixed or PTZ?

The Standard says cameras required to meet its defined Coverage Requirements should be static, and their field of view should not be changed by unauthorised users.

This is important because a PTZ camera can be pointing somewhere else at the exact moment an incident occurs.

PTZ cameras can still have useful surveillance roles, but they should not automatically replace a fixed camera required to maintain a specific view.

What does “fit for purpose” mean?

The final stage of the VSS process is not simply:

“The cameras are installed and online.”

The Standard says the building owner should verify after installation and commissioning that the deployed system:

meets the Operational Requirements and is fit for purpose.

That means asking whether the installed system actually performs the function that justified the camera in the first place.

Stage 4: How should CCTV be commissioned?

The Standard says agreed camera views and image quality for both:

live monitor views

and

recorded images

should be documented when the system is commissioned and reviewed periodically afterwards.

This is much stronger than checking:

“Camera online – PASS.”

A meaningful commissioning process should verify the actual image performance.

Why should the recorded image be checked separately?

Because what looks good on a live display may not necessarily produce the same level of useful detail after:

compression, recording and playback.

The Standard specifically warns that compression can reduce recorded picture detail.

Consultant’s perspective

A commissioning test should therefore ask:

Can we achieve the required surveillance outcome from the recorded evidence, not just from the live view?

That distinction matters greatly during investigations.

Does the VSS Standard specify how long CCTV footage should be retained?

Yes.

The Standard says the recording system should support continuous 24-hour recording of all VSS cameras and retain one full set of recordings covering the previous 31 days or more.

It also calls for at least 20% reserve recording media and minimum RAID 5 support.

We will treat storage architecture, RAID, reserve capacity and retention calculations properly in the dedicated recording article rather than compressing those topics here.

Does the VSS Standard specify frame rates?

Yes.

For day-to-day recordings, it states minimum recording rates of:

6 fps for indoor cameras

and

12 fps for outdoor monitoring of slow-moving traffic such as driveways.

It also says the capability to record selected cameras at 25 fps would be useful.

Again, the correct frame rate is not simply about choosing the largest number.

Frame rate affects:

storage, bandwidth and how movement appears in recorded video.

We will go much deeper into this in the recording article.

Does CCTV need backup power?

The Standard says VSS equipment such as VMS, NVR systems and viewing terminals should have UPS backup of at least 30 minutes.

The purpose is to maintain operation until generator backup becomes available or allow systems to shut down properly during an extended outage so recorded-image integrity can be preserved.

Should CCTV integrate with access control and intrusion alarms?

Yes, the Standard recognises the benefits of integrating VSS with other IP security systems such as:

access control, intrusion detection and video analytics.

It gives the example of an integrated system:

alerting security personnel, showing the alarm location on a site plan and automatically displaying the relevant camera for verification.

This converts CCTV from a passive recording system into a more useful operational tool.

How much spare capacity should a VSS have?

The Standard says VSS should be designed and installed with at least 20% spare capacity to support future expansion, with expansion causing minimum disruption to the existing system.

The detailed application of that 20% to:

channels, storage, switches, servers or other infrastructure

should be considered carefully during system design rather than casually assuming one interpretation for every component.

Does CCTV need maintenance after commissioning?

Absolutely.

The Standard says VSS should have a maintenance regime that keeps operational requirements consistently satisfied.

It gives a target system availability of 95% over a 12-month period and says equipment faults should be rectified within 24 hours, or sooner where the fault causes serious loss of coverage.

This means VSS compliance cannot be assessed only on installation day.

The system needs to continue performing.

Who should check that the CCTV system still works properly?

The Standard gives the building’s Security Manager responsibility for proper implementation of the VSS Operational Requirements.

This includes regular auditing of:

storage duration, alarms, live image quality and recorded image quality.

That is an important principle.

A camera being shown as online does not necessarily mean it still provides the required image.

A camera can remain online while:

its view is obstructed, focus has deteriorated, lighting conditions have changed or retention has fallen below the intended period.

Do CCTV operators need training?

Yes.

The Standard says VSS operators should receive suitable training and understand:

what to look for, how to operate the VSS and how to respond when an incident occurs.

It also recommends sufficient breaks, SOPs and regular refresher training.

This is another reason CCTV should not be treated merely as an equipment installation.

Its effectiveness depends partly on the people operating it.

Is cybersecurity part of physical CCTV design?

Yes.

The Standard includes cybersecurity as a dedicated consideration for IP-based VSS.

It recommends that building owners consider cybersecurity risk and controls against:

unauthorised access, interference and disabling of the VSS.

Its Annex B provides further recommendations around areas such as:

firmware updates, passwords, user accounts and secure communications.

We will cover these in detail in the VSS maintenance and cybersecurity article.

Why can a technically working CCTV system still be a poor security system?

Because individual components can work while the operational outcome fails.

For example:

A camera can be online.

The recorder can be recording.

The monitor can display the camera.

But if a person entering the building occupies too little of the image to be identified, then an Identification requirement has still not been achieved.

Likewise:

a perfectly recorded alarm event provides little operational value if nobody notices the alarm or knows what response is required.

This is why the Standard repeatedly focuses on the VSS as a complete system rather than individual products.

Practical VSS Design Framework

A useful way of applying the Standard is:

1. Define the problem
   What threat, vulnerability, safety issue or investigative need exists?
2. Decide what needs to be seen
   Person, vehicle, activity, doorway, asset or wider scene?
3. Define the required level of detail
   Detect, Observe, Recognise or Identify?
4. Establish where that requirement must be achieved
   At the doorway? Counter? Vehicle gate? Critical Asset?
5. Design the camera view
   Position, angle, focal length and field of view.
6. Design the environment around the camera
   Lighting, physical obstructions and tamper exposure.
7. Define recording requirements
   Resolution, frame rate, continuous recording and retention.
8. Design the infrastructure
   Network, storage, power, UPS and spare capacity.
9. Integrate the VSS where operationally useful
   Access control, intrusion alarms and analytics.
10. Define operator response
    Who sees the event and what happens next?
11. Commission against the Operational Requirement
    Prove that the installed system achieves the required view and recorded quality.
12. Maintain and periodically review it
    Because a system can degrade or its environment can change after installation.

This is a consultant’s implementation structure based on the VSS Standard rather than a checklist reproduced directly from the document.

Common misunderstandings about the VSS Standard

“CCTV design starts with calculating how many cameras the building needs.”
No. The Standard starts by defining the security problem and Operational Requirements.

“If the camera is 1080p or higher, the design complies everywhere.”
No. Resolution alone does not determine whether Detect, Observe, Recognise or Identify requirements are achieved.

“A 4K camera will always identify a person better than a 2MP camera.”
Not necessarily. Field of view, distance, lighting, angle and compression also affect usable detail.

“If the camera can see the entrance, the entrance is covered.”
Not necessarily. Main entrances require a frontal view meeting an Identification-level objective under the Standard.

“PTZ cameras can replace all fixed cameras.”
The Standard says cameras required to achieve defined Coverage Requirements should be static.

“CCTV is mainly an evidence system.”
Evidence is one function. The Standard also considers real-time detection, monitoring, response and incident management.

“Once the installer says every camera is online, commissioning is complete.”
No. Agreed views and image quality for both live and recorded images should be documented and periodically reviewed.

“Once CCTV has been installed, it should work indefinitely.”
No. The Standard expects maintenance, regular audits and timely fault rectification.

Conclusion

The SPF VSS Standard is best understood as an outcome-based approach to surveillance design.

It does not begin by asking:

“Which camera should we buy?”

It begins by asking:

“What problem needs to be solved?”

From there, the building owner should define the operational requirement, develop suitable technical specifications, install the system and finally verify that the system is genuinely fit for purpose.

That process links together:

camera placement → field of view → image detail → lighting → recording → storage → monitoring → integration → response → commissioning → maintenance.

The strongest CCTV design is therefore not necessarily the one with the highest resolution or the greatest number of cameras.

It is the one where:

each camera has a defined purpose, delivers the required image at the required location, records that evidence reliably, and supports an effective security operation.