Scanners and Optics Explained: Mirrors, Lens & CCD Array

[mhdeia]

Hello all!
I am really rconfused about the optical part of a scanner head.
I read that it consists of mirrors, a lens and the ccd array. But how this lens looks like? Is it like a large slice? Do really exist such lenses? How can such a lens have so good resolution and so large field of view (that the scanner needs to move only to one direction!)

Any information about the scanner's optics will be appreciated :-) because until now I have found almost nothing...

THX!

 

[davenn]

Hi there
welcome to PF :)

what do you mean my a "large slice" ?

a scanner lens particularly in something small like a handheld scanner is nothing more than a glass "D" shaped lens a few centimetres or so long

Light is shone down onto the paper to be scanned light is reflected from the paper and focussed by the lens into the long narrow CCD or photodiode array.
As this is scrolled across the page the sensor ( photodiodes/CCD strip) detect the variations in light and dark areas and it builds up a picture line by line

Optical character recognition software can then be used to detect text and display it appropriately

Dave

 

[davenn]

a quick drawing for you, something like this...

cheers
Dave

 

[mhdeia]

Thank you very much davenn
Large slice.. actually I mean like a rod, like the one that draw.
Well, it's a bit strange for me that you can have such a good quality and resolution with that type of lens, whereas when you want to see something smaller (~ 3um resolution) you have to use an expensive objective lens (with a lot of elements inside to correct aberrations etc)... Lenses for scanner don't suffer from aberrations and diffraction effects?

For example in laboratory we used the pathscan enabler IV scanner to see a microscope slide and it had resolution ~ 5 microns without the use of an objective lens.

 

[Drakkith]

No, they suffer from aberrations just like anything does. You just can't tell because you aren't magnifying anything like you are with a microscope or telescope. Also because the lens is a cylindrical lens (or something like that) and each pixel is at the optical axis of the lens. In a normal telescope or microscope you want to capture an area all at once, not have to scan the scope around to move the image across the pixels, so you have a square or rectangular chip. (With some exceptions) So in essence you actually have a very very tiny field of view, only 1 pixel across, but it is as long as the lens is. And it takes thousands of exposures each time you scan a page and combines them into one picture.