Re: To David Soderman: A Little Test

[Arthur Entlich <artistic-1@xxxxxxx>]

The Nikon scanners are unique in their design in that rather than 
providing color separating filters over the CCD element lines, they 
flash a sequence of LED lights RGBRGBRGB, or with dICE 
RGB(IR)RGB(IR)RGB(IR).  On their more recent scanners, they move the 
whole optical assembly rather than the film, so once the film is 
positioned in the correct frame, it works more like a flatbed, in that 
the scanner mechanism moves while the film remains stationary.

The Nikon LS-8000 scanner can either use one or three of these CCD 
tri-lines with a diffused colored LED light array.

Sequence of events: (this is more theoretical than actual, since I have 
never studied one)

Red LED lighting comes on, capture of one or three lines taken

Green LED goes on, capture of one or three lines taken

Blue Light goes on, capture of one or three lines taken

Optional IR goes on, capture of one or three lines taken

Each of these scans produces one or three lines of information, each 
color separated for that specific color of the lighting, making a 
monochromic (Black and white) image of 14? bits.

Eventually, all the needed lines of info are captured, and each capture 
of the specific color (like all lines captured when the RED LED array 
was on) are then amassed into three (or four with IR) separate channels 
of full images, each monochromatic, but referenced to and associated 
with either R, G, B, or IR.

Then combined, they make a full color image or 42 bits (14 bits per 
color channel - the IR channel isn't included in this), which is padded 
to 48 bits in something like Photoshop.

Just so you understand, all currently made CCD sensor elements (at least 
that I am aware of) are really monochrome, they do not generate color. 
They simply are measuring the light intensity once it is filtered 
through the source image and the color filter (in most scanners) or the 
other way around in the Nikon, (the light source is colored to begin 
with, and it goes through the image, and what comes out the other side 
is measure by the CCD sensor element.

You are right that the Nikon could have any number of sensor lines since 
each records all three colors scans and IR.  I imagine Nikon is using an 
sensor produced for color work but has it made without the colored 
filters on the three lines.

I hope this is somewhat helpful,

Art


Paul D. DeRocco wrote:

> I don't think it could be, because then it wouldn't be able to do superfine
> mode with a single line. It sounds like it's three lines, each of which has
> RGB sensors, whose only purpose is to be able to capture three scan lines at
> a time. I don't think the fact that there are three lines has anything to do
> with the fact that there are three primary colors--they could have used a
> four-line CCD and made it four times as fast as superfine mode.
> 
> Or maybe I'm missing something.
> 
> --
> 
> Ciao,               Paul D. DeRocco
> Paul                mailto:pderocco@ix.netcom.com
> 
> 
>>From: orafe.bustin@verizon.net
>>
>>On 9 Apr 2002 at 21:37, gary wrote:
>>
>>
>>>So it is tri-line but monochrome (i.e. no color mask), right?
>>>
>>yup, you got it.
>>
> 
> -
> Turn off HTML mail features. Keep quoted material short. Use accurate
> subject lines. http://www.leben.com/lists for list instructions.
> 
> 


-
Turn off HTML mail features. Keep quoted material short. Use accurate
subject lines. http://www.leben.com/lists for list instructions.