Re: How many pixels?

"Gary G" <see.signature@bottom> wrote in message
news:sq3e429itnv0erudkpelhrno9jjjpohik0@xxxxxxxxxx
On Thu, 20 Apr 2006 02:26:31 GMT, "NoSpam" <NoSpam@xxxxxxxxxxx> wrote:




What about final magnification at the camera sensor? I.e., if the
oculars are 10X and the objective is 100X, the final mag is 1,000X.
This assumes that the camera is parfocal with the oculars (which can
be done).


Gary Gaugler, Ph.D.
Microtechnics, Inc.
Granite Bay, CA 95746
916.791.8191
gary@microtechnics dot com

No, I am afraid a 10 times factor between the intermediate
image and that at the CCD may only serve to adjust the size
of the resolved elements in the image plane to the pixel size
on the CCD. It will not increase the number of resolved elements
on the object. This number is fixed once and for all by the
objective lens(and its illumination).

If the image formed by the objective consisting for example of
1000x1000 elements or pixels is accurately imaged onto a CCD
of say 1000x1000 elements then you have a perfect match.

If you enlarge the 1000x1000 elements in the exit pupil of the
objective so that they will cover an area 4 times larger than the
1000x1000CCD array, then each element resolved by the objective
will cast\ its light upon 2 CCD elements. 75 % of the image will
be lost. The remaining resolved elements will illuminate about
2 pixels with reduced energy flux but with the same color. This
will not enhance or diminish the resolution of the device but
decrease the portion of the object which is imaged. This portion
will then appear enlarged but will not be better resolved. This is
the case of empty magnification.

If the exit pupil is imaged upon a smaller number of pixels than
the 1000x1000 array covers, then one of these pixels will receive
more than one resolved element and the resolution achieved for
this arrangement will be decreased. All of the imaged area is
imaged, but at decreased resolution. In this case more magnifi-
cation will give a better resolved picture.

Now the question will arise: why should it be that a microscope
camera does not need all the many pixels which an ordinary
digital camera may offer? The reason is that the objective in the
ordinary digital camera has a linear resolution determined by the
same formula as the objective lens in a microscope (lambda/2xNA).
The linear size of the resolved element in the detector plane is
therefore the same for all cameras with a lens of the same f-number.
If the size of the detector array is larger, as it is for lenses of longer
focal length, then there are more resolved elements in the image
plane. An image of the same scene will be rendered at higher
resolution in the camera with the larger array or longer focal
length. (This of course supposes that the technology of the
CCD array is comparable for the cameras being compared.)
There is no intermediate image and no second magnification
involved here.

One could enhance the number of useful pixels for a micros-
cope in a similar fashion. One could possibly construct
objective lenses at any of the current NAs with longer focal
lengths. All of these objectives will resolve at identical
linear distance. But the ones with the longer focal lengths will
have larger intermediate images and at the same linear resolution
would allow the use of CCDs with an increasing number of
useable pixels. The disadvante is that for quadrupling
the number of resolved pixels one would have to build micros-
copes twice the size of the present scopes and about four
times their weight.

G.R.











.

Relevant Pages

  • Re: DPI
    ... The pixel originates in your camera. ... There are roughly eight million Pixels on the sensor in your camera. ... resolution discernable by the human eye. ... Printer resolution is how many dots of ink the ...
    (alt.photography)
  • Re: How many pixels?
    ... This assumes that the camera is parfocal with the oculars (which can ... on the CCD. ... It will not increase the number of resolved elements ... 1000x1000 elements or pixels is accurately imaged onto a CCD ...
    (sci.techniques.microscopy)
  • Re: Is 16x9 really 16x9?
    ... > NTSC DV video is 720 x 480. ... > camera technologies have square pixels, ... > resolution assumes rectangular pixels. ...
    (rec.video.desktop)
  • Re: DPI
    ... I have read that cameras work on 72 dpi ... The pixel originates in your camera. ... There are roughly eight million Pixels on the sensor in your camera. ... resolution discernable by the human eye. ...
    (alt.photography)
  • Re: Nikon Coolpix with swiveling lense?
    ... 452 horizontal pixels ... 336 vertical pixels ... This isn't even up to the resolution in an extra 640x480 image border on only 2 ... resolution of a 5mpx camera you'd have to go up to a 20mpx camera. ...
    (rec.photo.digital)