magnification of digital microscopes

[Andreas Bruckbauer]

Hello,
somebody asked me about the magnification of our TIRF microscpe, however
this should also apply to confocal systems. I came up with the following:
The optical resolution of a high NA system is about 0.2 micrometer and
applying 2x Nyquist sampling this is about 100 nm for one pixel (on CCD or
confocal). At the end of the image processing pipeline this will be printed
or displayed somewhere at a resolution visible to the eye. This should be
one line pair per 1/60th degree which is about 70 micrometer at 25 cm
viewing distance. Gong from line pairs to pixels would give 35 micrometer
and the magnification would then be 35/0.1 = 350 x (rather low). Of course
one could print it bigger and on a compter screen it would be much bigger
but this would be empty magnification. Is this right? Why is useful
magnification for optical microscopes given as 500 x NA which would be 700x
in the above case? Is this because the fluorescence is so dimm that the eye
achieves lower resolution?

[Guy Cox]

The conventional LM recommendation is based on direct viewing through the eyepiece.  The human eye is assumed to have a resolution of 200µm, so to get 200nm to 200µm is a magnification of x1000.  This is about 700 x NA.  You get roughly the same with a x40 NA 0.65 objective, which has a resolution of 500nm so you need a magnification at the eye of x400.  

If you are making a print on paper at a magnification of x350 your 200nm resolution will come out at 70µm, as you say (there's really no need to get pixels involved).  This is way below the commonly used value for resolution of the eye, hence the lower magnification.  Maybe some very young people with perfect eyesight could get to this value.  I've not met many people who can see the mesh of a 400 lpi EM grid (63.5µm) though young folk can often see 200 mesh (127 µm).  

So the brief answer would seem to be that you are taking a rather optimistic value for the resolution of the eye - where did you get that figure from?

                                         Guy

Optical Imaging Techniques in Cell Biology
by Guy Cox    CRC Press / Taylor & Francis
     http://www.guycox.com/optical.htm
______________________________________________
Associate Professor Guy Cox, MA, DPhil(Oxon)
Electron Microscope Unit, Madsen Building F09,
University of Sydney, NSW 2006
______________________________________________
Phone [phone]     Fax [phone]
Mobile [phone]
______________________________________________
      http://www.guycox.net
 
-----Original Message-----
From: Confocal Microscopy List [mailto:[hidden email]] On Behalf Of Andreas Bruckbauer
Sent: Thursday, November 12, 2009 8:58 AM
To: [hidden email]
Subject: magnification of digital microscopes

Hello,
somebody asked me about the magnification of our TIRF microscpe, however
this should also apply to confocal systems. I came up with the following:
The optical resolution of a high NA system is about 0.2 micrometer and
applying 2x Nyquist sampling this is about 100 nm for one pixel (on CCD or
confocal). At the end of the image processing pipeline this will be printed
or displayed somewhere at a resolution visible to the eye. This should be
one line pair per 1/60th degree which is about 70 micrometer at 25 cm
viewing distance. Gong from line pairs to pixels would give 35 micrometer
and the magnification would then be 35/0.1 = 350 x (rather low). Of course
one could print it bigger and on a compter screen it would be much bigger
but this would be empty magnification. Is this right? Why is useful
magnification for optical microscopes given as 500 x NA which would be 700x
in the above case? Is this because the fluorescence is so dimm that the eye
achieves lower resolution?

No virus found in this incoming message.
Checked by AVG - www.avg.com
Version: 9.0.702 / Virus Database: 270.14.50/2481 - Release Date: 11/11/09 06:56:00

[Andreas Bruckbauer]

Many thanks for the reply, I got the 70 µm from the 1 arc-min resolution for 20/20 vision as given in the Handbook of biological confocal microscopy by Pawley (Chapter the pixelated image). However I wrongly assumed that this would be for a line pair. It seems to be specified for the thickness of a line in the testpattern which is used for eye testing http://webvision.med.utah.edu/KallSpatial.html#introduction. So for one line pair it would be 140 µm. The EM grid has relatively thin lines compared to the test pattern so it might be more difficult to see and he mesh value would also correspond to a line pair. What would one need on the microscope side to make a fair comparison? The Rayleigh limit seems quite arbitrary.

Andreas

-----Original Message-----
From: Guy Cox <[hidden email]>
To: [hidden email]
Sent: Thu, 12 Nov 2009 3:37
Subject: Re: magnification of digital microscopes

The conventional LM recommendation is based on direct viewing through the 

eyepiece.  The human eye is assumed to have a resolution of 200µm, so to get 

200nm to 200µm is a magnification of x1000.  This is about 700 x NA.  You get 

roughly the same with a x40 NA 0.65 objective, which has a resolution of 500nm 

so you need a magnification at the eye of x400.  



If you are making a print on paper at a magnification of x350 your 200nm 

resolution will come out at 70µm, as you say (there's really no need to get 

pixels involved).  This is way below the commonly used value for resolution of 

the eye, hence the lower magnification.  Maybe some very young people with 

perfect eyesight could get to this value.  I've not met many people who can see 

the mesh of a 400 lpi EM grid (63.5µm) though young folk can often see 200 mesh 

(127 µm).  



So the brief answer would seem to be that you are taking a rather optimistic 

value for the resolution of the eye - where did you get that figure from?



                                         Guy 



Optical Imaging Techniques in Cell Biology

by Guy Cox    CRC Press / Taylor & Francis

     http://www.guycox.com/optical.htm

______________________________________________

Associate Professor Guy Cox, MA, DPhil(Oxon)

Electron Microscope Unit, Madsen Building F09,

University of Sydney, NSW 2006

______________________________________________

Phone [phone]     Fax [phone]

Mobile [phone]

______________________________________________

      http://www.guycox.net

 

-----Original Message-----

From: Confocal Microscopy List [[hidden email]] On 

Behalf Of Andreas Bruckbauer

Sent: Thursday, November 12, 2009 8:58 AM

To: [hidden email]

Subject: magnification of digital microscopes



Hello,

somebody asked me about the magnification of our TIRF microscpe, however

this should also apply to confocal systems. I came up with the following:

The optical resolution of a high NA system is about 0.2 micrometer and

applying 2x Nyquist sampling this is about 100 nm for one pixel (on CCD or

confocal). At the end of the image processing pipeline this will be printed

or displayed somewhere at a resolution visible to the eye. This should be

one line pair per 1/60th degree which is about 70 micrometer at 25 cm

viewing distance. Gong from line pairs to pixels would give 35 micrometer

and the magnification would then be 35/0.1 = 350 x (rather low). Of course

one could print it bigger and on a compter screen it would be much bigger

but this would be empty magnification. Is this right? Why is useful

magnification for optical microscopes given as 500 x NA which would be 700x

in the above case? Is this because the fluorescence is so dimm that the eye

achieves lower resolution? 



No virus found in this incoming message.

Checked by AVG - www.avg.com 

Version: 9.0.702 / Virus Database: 270.14.50/2481 - Release Date: 11/11/09 

06:56:00

[Michal Opas]

Well, the List is my last hope.
The problem:
after refurbishing Kr/Ar laser and re-installing it I am not able to get a sharp image on the monitor. My gut feeling is that it probably is not related to laser but a problem with a scan head. So laser outputs fine, staining (as excited by Hg burner) looks spectacular but to get an output pinhole has to be open and gain cranked up. ND filter is on 1. And the image still look blurry.
Any advice as to how to approach the problem shall be greatly appreciated.

Michal

[W. Chan]

Hi Michael

If it is a new laser tube, there must be something wrong that you have to
use ND position 1 which is 10% transmittance, in addition to a large
pinhole and high gain.  I suspect it's an alignment problem with the laser
coming into the scanhead.  What's the anode current, power measured at the
laser head and at the stage (specimen level)?

If all those are good, then we can try trace along the light path: filter
blocks and other mirrors in the scanhead...

On Fri, 4 Jun 2010, Michal Opas wrote:

> Well, the List is my last hope.
> The problem:
> after refurbishing Kr/Ar laser and re-installing it I am not able to get a
> sharp image on the monitor. My gut feeling is that it probably is not
> related to laser but a problem with a scan head. So laser outputs fine,
> staining (as excited by Hg burner) looks spectacular but to get an output
> pinhole has to be open and gain cranked up. ND filter is on 1. And the image
> still look blurry.
> Any advice as to how to approach the problem shall be greatly appreciated.
>
> Michal

--
Pang (Wai Pang Chan, [hidden email], PAB A087, 206-685-1519)
The Biology Imaging Facility (http://depts.washington.edu/if/)

[Mark Cannell]

Sounds like the alignment is off, not surprising after laser refurbishing.

If you remove the objective with the scan mirrors stationary in the
center of the field, what does the rear aperture illumination look like?
If that is OK, use a mirror as a reflector and examine the alignment on
each folding mirror.

Hope this helps

Mark



Michal Opas wrote:

> Well, the List is my last hope.
> The problem:
> after refurbishing Kr/Ar laser and re-installing it I am not able to
> get a sharp image on the monitor. My gut feeling is that it probably
> is not related to laser but a problem with a scan head. So laser
> outputs fine, staining (as excited by Hg burner) looks spectacular but
> to get an output pinhole has to be open and gain cranked up. ND filter
> is on 1. And the image still look blurry.
> Any advice as to how to approach the problem shall be greatly appreciated.
>
> Michal
>
> --
>
> *9^th International Calreticulin Workshop will be held in *
> *Copenhagen* *, * *Denmark* * on * *August 29^th -31^st 2011* *. For
> additional information please visit:** http://www.crt2011.com/ *
>
>
>      * Dr. Michal Opas
>      Professor
>      Department of Laboratory Medicine and Pathobiology
>      University of Toronto
>      1 King's College Circle
>      Medical Sciences Building, room 6326
>      Toronto, Ontario, M5S 1A8 Canada
>
> **°°°°°°°°°°°°°**
>  phone: [phone] (laboratory)
>          [phone] (office)
>    fax: [phone]
>  e*-*mail: **[hidden email] <mailto:[hidden email]>** *
>
> *WWW* *:** **http://www.utoronto.ca/mocell
> <http://www.utoronto.ca/mocell>*
>