Video Security System Design Considerations
Updated: Apr 10, 2019
Whenever I visit the site of a new client who is interested in protecting their property with video security, the first question I always ask is “what are you trying to achieve?” More often than not, the client does not understand what I am asking so I happily spend some time to explain. One of the reasons for writing this article will be that it will become an introduction to the issues surrounding video security and a first point of reference before I visit. This will help to give the client some background on what video can do and probably save some time and avoid confusion.
There’s a common misconception that all video security systems are pretty much the same and all you have to do is slap a few cameras, from a hardware store, on the wall or ceiling and everything will be covered. This is not the case. A properly designed system based on the needs of the client will make all the difference to the end result.
One of the main factors in deploying a video system is understanding what asset is to be protected and what level of surveillance is required. Broadly speaking, a video system should do one or more of the following:
Observe an area or target.
Recognise a person or object
Identify an unknown person or object
Track and document a person or event.
So, what do I mean by each of these and how does it affect the system design?
A standard video security camera will generally have a 3.5mm, or thereabouts, lens. This is roughly the field of view that you would get if looked at an area with one eye closed. It’s roughly a 90 degree view of the area and is not specifically focused on anything. This view allows you to adequately observe movement, such as a car driving through a car park, but doesn’t give you any useful detail unless the target is within a few metres of the camera. From a distance, you may see that a person gets into a red car and drives off but you probably won’t be able to tell who it is or obtain any detail of the car. It’s perfect for “keeping an eye on things” in a large area and allows you to track where a person goes or what happens to an object, such as a car.
Being able to recognise a person or object requires a higher level of detail. This means that the camera might need to be closer to the object or have a longer focal length. The effect would be similar to zooming with your pocket camera. Either way, the target area covered will be smaller but the detail of what is covered will be higher. This will allow you to recognise someone or something that you already know. For example, if you are familiar with Steve and he moves through the field of vision, based on his overall appearance, physical characteristics, clothing or other distinct feature, you will know that it’s Steve. There will be insufficient detail to recognise him based on facial features. If you only have a couple of cameras covering smaller areas, you won’t be able to track what he does or where he goes and you may not be able to prove that it’s Steve if required.
If an event occurs, let’s say a theft or assault, you may need to have a clear image of a person that you do not know. Fuzzy images taken from a distance will not be sufficient to assist police or security offices to identify the culprit. For this purpose you will need a camera which is tightly focused on a target area and has sufficient resolution to be able to provide a clear image of distinctive features, such as a face. To be able to obtain this image, a camera will need to be placed at location and at a height which will maximise the probability of capturing this level of detail. We commonly refer to such an area as a “choke point”. This is a common point through which people must pass and will often be an entry or exit point or a common walk way.
I also mentioned camera height when considering placement in a choke point. It is no use having the camera mounted on or near a ceiling as the most likely image captured will be the top of someone’s head or, worse still, a hat, thereby making them unidentifiable.
As an example of a choke point which I have found to be quite effective, I like to mount a camera on a wall at the top of a set of stairs, pointing down the stairs. Human nature, and my own experience in camera placement, seems to indicate that people who walk up stairs will look up as they approach the top of the flight; probably to see who is around or where they have to go next. This affords a great opportunity to capture a really clear face shot. Of course, anyone intending mischief and who is aware of the presence of the camera will deliberately NOT look up. This is where concealed cameras come in but that’s a story for another time.
So far, I have identified a number of factors which are important in determining where cameras should be placed and, together, these factors contribute to achieving the outcome required by the client.
At the end of the day, most clients want their shiny new system to do everything I’ve identified and this is why a layered approach is required. Using the scenario of a shopping centre car park, a typical camera placement strategy may include:
One or more wide angle cameras covering most of the car park from different angles to detect changes and to track movement
One or more cameras with higher resolution and longer focal lengths to capture more specific detail with minimal graininess or pixilation when zooming the image. This may also be achieved by having cameras closer to the target asset under observation, such as the lift areas or retail entrances. Identity recognition will require higher resolution cameras.
Highly focused, high resolution cameras targeting the car park entry and exit points allowing for vehicle and number plate identification. Cameras for this application need to cope with variations in lighting and target movement. A camera with a Wide Dynamic Range (WDR) and a fast shutter speed will often be required.
Using this scenario, a car entering the car park can be identified. It can be tracked as it moves through the car park to where it stops. The driver or occupants can be tracked once they leave the vehicle and they can be identified as they enter and leave the retail section.
In addition to issues relating to camera placement, it is critical that cameras cannot be taken out of service without the intruder first being identified by one or more other cameras. For this purpose, cameras should not be accessible from the rear and any connecting data and power cables must be adequately protected. Proper design and installation by a qualified technician is essential to maintaining the integrity of the system.
Finally, for the purpose of law enforcement, the images captured by the cameras need to be stored on a suitable recording device and retained for a mandatory period of time. This will either be determined by regulation or client preference. The specification for this device will be determined by the number of cameras, the amount of data traffic generated and the time for which the video must be stored.
There are other issues relating to the installation of video security cameras but the purpose of this article is to highlight the issues confronting the system design with respect to the client’s requirements and the added value of engaging an experienced professional. A good designer will take the time to understand the client’s requirements, raise relevant issues and help the client to maximise the return on their expenditure.
If you are thinking of engaging a professional to install a video security system, ensure that they ask questions about your objectives. Ask to see, and have them explain, some other project designs in which they have been involved and, if you can, check the quality of the finish.
Author: Craig Colnan