Tuesday, September 11, 2007

Computer Vision Basics -cameras

" Every man is ultimately a plagarist. How on earth can we call anything as an original creation?" -unknown

To understand computational cameras needed for optical vision, one needs to have a good solid technical foundation on some optical nuances, some of which are gonna be covered below. I have not written up anything which is below . They are just excerpts form internet or wikipedia.

Whats the "entrance pupil" ? An entrance pupil in optical systems is the area at the entry of the system which permits the entry of light into the system. It is a kind of an aperture. An aperture controls the amount of light entering the lens. For a camera a smaller aperture usually results in a larger field of view . This means that objects far away form it would be in focus. A larger aperture would require more expensive lenses and imaging stuff for good images. A larger aperture will have a lesser field of view. You can try that in your own camera by using an f/2.8 and f/32 aperture.

The amount of light captured by a lens is proportional to the area of the aperture, equal to:

\mathrm{Area} = \pi \left({f \over 2N}\right)^2

Where f is focal length and N is the f number.



First on the list we need to know whats the f-number.
To put it in the easiest way, f-number is just simple: if the focal length is 5 times the diameter, the f-number is f/5.

No comments: