There are lots of different cameras available on the market. To decide which one and where: you need to consider your lens options, the resolution required, Frames Per Second and storage.
Most cameras will be static and aiming at a specific area. Below is a quick guide for lens options.
Other things to consider are the cameras aperture, exposure and shutter speed. The area being recorded may be inside in which case a Fisheye lens or 360 camera maybe a better choice.
Resolution – Most people enjoy watching TV programs where a simple ‘enhance’ lets them zoom in and clearly read someone’s shopping list. But in reality this isn’t possible unless you have a camera recording in a huge resolution. The problem with this is the camera will be expensive and storing the footage will be expensive also.
Frame Per Second (FPS) – When choosing the FPS a camera is recording at it is important to take into other considerations such as storage and bandwidth. In an ideal world you would run at the highest FPS on all camera but this can create large network usage and require large amounts of storage for a camera that is recording empty space 23 hours a day. We have found that between 8 and 12 FPS can produce smooth (enough) playback for monitoring of people but this is dependent on the situation as recording a road may require higher FPS.
The above are the basics to prepare for but you also need to have a good understanding of:
The area – how far away do you want the camera to see?
The detail – how much zooming in do you want to be able to do?
The motion – are you recording fast objects or just someone walking?
Codec – H.264 / H.265 / MJPEG – You will most likely be going with H.264, this will give you good quality images and use less bandwidth which will allow you to have more cameras on the network.
As well as the cameras codec you have to consider how the camera will connect to the system. works with most camera manufactures to create a custom connection to get the best results from the cameras. However, lots of cameras are designed to be generic so they can work with the most amount of systems, to do this they use an Onvif method. Depending on how well this has been implemented on the camera can be the difference between a high quality camera and a cheap camera. If quality is important then the custom connections always yields better results than the ONVIF.
Below are some other things to be aware of, but these things are normally automatic, and out of control of the camera and the NVR.
Transport Method – In networking there can be many different network paths used. With cameras and the high amount of network traffic they use the preferred method is UDP. Most other methods such as web browsing (even if its to a cameras IP address) will use TCP (RTSP).
RTP over RTSP – This is your normal TCP data and isn’t used by many VMS due to increase overheads in processing which can add lag to video footage; instead UDP is preferred. RTSP is very robust so very useful on networks with high packet loss.
RTP over UDP – UDP has less network overheads associated with it which allows it to run smoother, faster and with less processing. However, UDP is more fragile than RTSP so can suffer on networks with high packet loss.
Variable Bit Rate (VBR) – Can save on bandwidth during times of low activity but also uses greater bandwidth during high activity.
Constant Bit Rate (CBR) – Allows you to be more precise and calculating in your bandwidth calculations but comes with the down side of not getting any savings during low activity times.