As discussed in the last column, resolution is a very important part of the overall image. Regardless of being an IP or an analog image, if the clarity of the image is too little to determine identity of the object, person, area or whatever it is that you are watching, you may as well save your efforts and not install the system. The next important issues surround what lenses you will choose.
Choosing the proper lens for your application is equal to all other portions combined. If you end up with a view that is too wide, the best resolution camera on the market will not give you a clear image of objects that are too far from the camera. Equally, if you choose too much magnification, the scene will be very narrow and you may be required to use more cameras than you intended to cover the application.
The first step to choosing lenses is to understand that it the decision is 100 percent based upon mathematics and therefore not magic or luck.
Starting at the beginning, we will assume that we are working with the defined camera application. You have your viewing position, purpose and required criteria worked out. We will assume that you have your camera style worked out as well -- that is, what type of camera you will need according to the light levels that you have to work with: day/night, color, low light, etc.
I will not assume that you have decided upon a standard camera or a mega-pixel camera yet, but you should at least have an idea as to whether or not you plan to use a fixed or Pan/Tilt (PT). Right off the bat, if you are planning to use a PT, you will most likely be looking to use a zoom lens, so you will calculate two ratios of lens ... the wide and the telephoto. However, with the IP solution, you may be able to replace the PTZ idea with a fixed mega-pixel camera. Again, it is determined by your application and transmission capabilities.
Let's start with the fixed lens. In order to calculate what lens to use, you must have three points or criteria. You must know the format size of the camera, the distance between the camera and the scene to be viewed, the width of the scene that you wish to view or the height of the scene that you wish to view. Without three of the four mentioned criteria, you cannot calculate a lens.
The width or height of the scene will be based upon what you are trying to see. The size of the objects required in your view for identification or viewing will also matter. Based upon general applications, the object of view should take up at least 10 to 20 percent of the overall width of the scene. The smaller the detail of the object, the more of the scene you must have focused on the object. Personal identification of a human should be 25% of the overall scene width as based upon the SWGIT CCTV guidelines (Scientific Working Group on Imaging Technology; recently adopted by the FBI as a standard on CCTV system design).
From here, there are four or five different methods for calculating a lens. You can use a formula:
SCENE WIDTH = (Format (Hz) X Distance) / Focal Length
For this, you need to know certain statistics about the imager in the camera that you are using, i.e. the horizontal or vertical format size of the imager. The second (and easier) method would be to use a field of view calculator. These wheels or slide rules are available, for free, from almost all of the major manufacturers and are very easy to use.
You could also use a cheat sheet. This is a form that list several different sizes of lens down the left column and general distances in feet or meters across the top. The objective is to match the distance with the lens and come to the point where they cross and see the field of view that you would create.