For once, let’s put camera lenses right at the top of the list. They are absolutely vital to the picture results that you will achieve. Their performance is key in controlling the amount of light that actually reaches the CMOS or CCD imaging chip in the camera, and the optical quality will determine just how clear (or high-resolution) an image you are able to achieve.
Furthermore, the lens will determine exactly what field of view (FOV) your camera is able to cover, and the size of the scene which you try to cover is absolutely essential to the level of detail that you will be able to see within the images..
OK, there’s a bit of ground to cover and explain, so let’s break it down ito bite-size pieces:.
Auto-Iris (AI) or Direct Drive (DD) Lenses
The lens iris is the adjustable aperture (hole) at the front of the lens which is adjusted to vary the amount of light passing through the lens to reach the camera’s imaging chip.
If you are installing cameras to monitor any scene where the light levels are likely to vary, then you should be using AI or DD lenses. Just about the only places where light levels do not vary are windowless rooms!
Fixed-Iris lenses are cheaper, and for this reason alone you will see them used fairly often.
Obviously, if light levels vary, it is better to be able to adjust the iris to allow only a useful amount of light to fall upon the imaging chip.
With an auto-iris lens the camera and lens are connected by a cable and varying light levels are automatically compensated for; the iris opens if more light is needed, or closes if too much light is reaching the chip.
Field Of View (FOV)
The focal length of the lens determines the field of view which the camera will be able to cover.
As an indication, a lens with a low value focal length, say 3mm, will have a wide field of view, perhaps 90°.
Transversely, a lens with a high value focal length, say 50mm, will have a narrow field of view, perhaps just 5°.
For best results it is vital that you select a lens which is able to closely cover only the field of view (or scene if you prefer) that you need to cover.
Bear in mind that your camera and lens can only deliver a fixed aspect ratio – currently 4:3 format is still the most common format for the types of cameras we are discussing. This means that for every increase in scene width that you try to cover there will inevitably be a corresponding increase in scene height. Put simply, the wider that you make the view the shorter and smaller people will appear within the scene.
A few years ago it was necessary to calculate the lens focal length which was the closest fit to monitor the required scene. Nowadays vari-focal lenses have become very cost-effective and are therefore the default choice for all new applications. The vari-focal lens is adjustable between two values e.g. 3-9mm and enables you to exactly adjust (or ‘clip’) the view to only the area which you are interested in.
To use an analogy with hand-held digital cameras, what we are talking about here is optical zoom. If you want to achieve great results you use your camera’s optical zoom function to zoom-in on the terget area before depressing the shutter and taking the picture. You could actually take capture the image of a larger scene and then use software to clip the actual area of interest, but hopefully you will realise and understand that you are wasting a proportion of the camera’s resolution capability.
Transmission of Light (F-Stop)
The ability of a lens to allow light to pass through to the imaging chip is indicated by the F-stop value. The lower the value, the more light the lens allows through. F1.0 is a good value, some lenses with a high focal length value (narrow field of view) may have a rating of perhaps F1.8. As an indication of the significance of this value, each whole stop (say from F1.0 to F1.2) indicates only half as much light getting through.
A fairly recent development is the aspherical lens which is specially shaped to capture as much light as possible and achieve a low F-stop rating.
Night-time and Infra-Red (IR) lighting
Plainly, the best starting point is to use a lens with a low F-stop, one which is able to allow as much light as possible through the lens to fall upon the imaging chip.
Sometimes additional artificial lighting will be needed. You should always consider white lighting first. White lighting will allow the camera to render true-to-life colour images, is a visible deterrent by illuminating the scene and is helpful to users of the space. Yellow (SON or sodium type) lighting will cause a yellow colour cast effect within the resulting images.
Infra-red lighting is often selected for night-time usage because it is invisible to the human eye. This means that the camera is able to monitor an area and provide useful (black < white only) images, but there is no light pollution impact upon the scene.
If you must use IR lighting, make sure that the lenses selected are optically corrected for IR lighting. IR light is at a different wavelength to white light and this may result in a focus shift between day < night images. Some inferior lenses are merely coated to correct for IR lighting at a pre-determined wavelength, this is not as effective as optically corrected lenses.
This is becoming ever more important with the development of megapixel cameras. Obviously, there is little point in paying for an expensive megapixel camera capable of delivering a very high resolution image and then fitting a cheap lens which will not allow the camera to resolve the required detail. Be sure to select a lens capable of matching the performance capability of the camera.
There’s more to lenses than meets the eye!
The points explained above are provided to assist you with lens selection. There are even more parameters and physical design matters which could be explained. Suffice to say that lenses are vital to system performance. These points should help you to understand why. The lenses presented in our web shop have been pre-selected to deliver best performance. You will need to choose the appropriate lens for your requirement.
Megapixel IP CCTV Cameras
The megapixel value is simply a shorthand means of expressing the camera’s image resolution in a single figure.
Multiply the horizontal resolution by the vertical resolution to arrive at a value which expresses the total number of pixels that the image device can deliver.
Because this multiplication, typically of thousands times thousands, results in such a large value we use the term mega (indicating 10 to the power of six, which is 1,000,000) to express the result.
You will often hear the term resolution referred to when comparing video products, so let’s have a look at what it means:
The most common method of stating resolution is to refer to it in pixel values e.g. 640 X 480.
The first value indicates the number of pixels across the image width, and the second value indicates the number of pixels in the image height.
Straight away you will be able to see that this method of expressing resolution also indicates the picture format. The two most common formats are 4:3 and 16:9 (or widescreen).
If you divide the first value in the example above (640) by 4 you get the result = 160; multiply this by 3 and you can confirm that such an image will be presented in the standard 4:3 format.
Essentially, the more pixels the better!
Despite what you may have seen on TV Spooks and Spy programs it is not possible to zoom-in, enhance, or enlarge images to uncover data that was never seen by the camera.
You need to bear in mind that this number of pixels is all you have to interpret your real world scene. So, if you fit a lens to a camera with an imaging device that has 640 pixels across the chip and set it up to view an image 6.4m wide, you will have only 1 pixel per cm. Believe me, the resultant image will be virtually useless.
The most common image formats are 4:3 and 16:9.
4:3 format is what all domestic TVs were for many years past.
16:9 is what we have come to know as widescreen (as used in modern LCD & plasma TVs and computer monitors).
The first figure refers to the image width.
The second figure refers to the image height.
Plainly these values are expressed as a ratio, meaning that (in the case of a 4:3 monitor) if you measure the monitor width and divide it by four, then it’s height will be three times that resulting value. It is a fixed ratio; the scene height will always be this proportion of the scene width.
You always need to bear this in mind when considering any particular scene to be monitored with a camera. The format ratio is fixed i.e. if you decide to view a wide scene, you will by default also capture a tall scene – it is impossible not to!
Wide scenes = tall scenes = small people in the scene = a lack of people detail.
We’ll talk more about this in our explanation of ‘Field of View’ (FOV).
The format that you actually use is determined by the image chip in the camera, it cannot be adjusted or changed.
Ideally, you would aim to view the images on a monitor with a matching format, and therefore endeavour to ensure that all cameras on a system are the same format.
Ethernet networks have been around for quite some time now.
We all expect our laptops and PCs to be manufactured with an ethernet port. Lately it has become the norm for our TVs, Hi-Fis, games consoles, and even kitchen appliances to be fitted with an ethernet port!
It is via a connection to this port that we are able to connect our machines to the network, and then onwards to the world wide web.
We have become reliant upon these connections for our broadband access to the internet, to surf the web and to collect our email. Furthermore, in the IT sector we accept it as the norm that we can share machines and data using such a network; we print to remote printers, and we might save our data on a hard drive on a remote machine or server.
It is therefore simply a logical progression of this which finds IP cameras and other peripheral devices becoming available for simple connection to these ethernet networks.
As with all things IT the speed of the network has increased steadily over time.
In ethernet terms we call this speed (or capability to handle data traffic quickly), the network’s bandwidth.
When PCs were first connected together in offices about 20 years ago they used a co-axial network cable which ran from machine to machine in a peer-to-peer manner and had a maximum throughput rate of 10 megabits per second.
That early co-ax network cable was fairly quickly replaced by what has become the ubiquitous cat5 network cable. cat5 cable contains 4 sets of twisted-pairs. Generally a pair is used to transmit the data, and another pair is used to receive the data (the other pairs are normally spare, but we’ll talk about uses for those when we discuss Power Over Ethernet (POE) in a separate article.
The network hardware devices (hubs & switches) were soon improved and developed to perform better with this new twisted-pair cabling, and as a result the network bandwidth was readily increased from 10 megabits per second to 100mbps.
From about 2007 cost-effective gigabit network switches became available, and it was simply a matter of replacing the hubs and switches in the network to achieve this further boost in bandwidth – up to 1,000mbps (or 1gigabit per second).
Further increases to a bandwidth of 10 gbps are already possible, but this is still quite demanding technology requiring special cable, shorter runs, etc.
You may also come across the term of reference BASE-T when discussing network bandwidth. Put simply:
10 BASE-T = 10 megabits per second
100 BASE-T = 100 mbps
1000 BASE-T = 1gbps
Good networks are vital to good system performance. Most large premises will have a specialist network engineer to manage the design, growth, balance and upkeep of their network.
IP is an acronym for Internet Protocol.
It works in combination with TCP, which is an acronym for Transmission Control Protocol.
Together they are known as TCP/IP.
Every network-connected device contains software to enable it to work with the TCP/IP protocols.
These protocols have been in common use for many years now. It is a stable and proven technology. The software within the devices is able to detect errors, missing or even corrupted information, and to manage the re-transmission of that data when necessary.
Each device on any network will be granted an IP address as a member of that network – an example of a typical form of IP address might be 192.168.1.5 Other devices in the same Local Area Network (LAN) would typically be granted IP addresses in the same range, this might be anything from 192.168.1.1 to 192.168.1.255.
Typically a single device on the network, such as a router or a server, will control the issuing of IP addresses to devices connected to the network. This is known as Dynamic Hosting, and will follow the Dynamic Host Control Protocol (DHCP) rules.
TCP/IP is now the universally accepted method of addressing and connecting to network devices, whether those devices be computers, card readers, cameras, telephones, fridges or fruit machines!
Using cameras as an example, IP networks typically allow multiple users (at their own client PCs on the network) to access and view multiple cameras connected at many other points on that network.
Perhaps the greatest strength of the Internet Protocol is the sheer fact that networks running this technology now reach everywhere in the world!
We are now able to add IP cameras or other devices to IP networks, and with the correct permissions we can access them from virtually anywhere.