I don't think that is quite true either. I think some aberrations increase and others decrease with distance. I have seen this when using a reversed camera lens on a tube lens. It may apply to microscope objectives too. This may be more important when using a wider tube lens, so that you will be pushing the limits of the field size. With a 200mm tube lens, maybe it really doesn't matter much."I understand the distance does not matter unless it is too long to cause vignetting."
Perfect distance between Sigma LSA and microscope lenses=?
Moderators: rjlittlefield, ChrisR, Chris S., Pau
Hi Lou,
Independent of the calibration of my system the specified range by NIKON: 100mm-200mm is for the LSA not really better then 10mm-50mm.
So, I will use the short one.
BR, ADi
Therefore I have been looking for the perfect distance :-) (especially for the focus-stacking!)”I think some aberrations increase and others decrease with distance”
Independent of the calibration of my system the specified range by NIKON: 100mm-200mm is for the LSA not really better then 10mm-50mm.
So, I will use the short one.
BR, ADi
- rjlittlefield
- Site Admin
- Posts: 24018
- Joined: Tue Aug 01, 2006 8:34 am
- Location: Richland, Washington State, USA
- Contact:
Yes, definitely.Lou Jost wrote:I think some aberrations increase and others decrease with distance.
On the general issue of aberrations and separation, you may be interested in the test results shown at http://www.photomacrography.net/forum/v ... 456#126456 . That test was set up to use a Raynox 150 as if it were a tube lens, but substituting just an 11mm aperture in place of an objective, and imaging a target located across the street at close to "infinity". So all that was in play there was the Raynox, the aperture, and the variable separation between those. Nonetheless, corner quality changed dramatically from minimal separation to 28 mm, and slowly between there and 84 mm (tested only at minimum, 28, 56, and 84).
--Rik
Hello everybody,
Does anybody know how to use the MTF mapper?
I have just tried with some test-pictures but the MTF-mapper has not recognized them up to now
Probably this program expects the whole test-picture. The quality of the whole test-picture scaled down is bad.
With the part of the test-picture it doesn’t work:
e.g. LSA & LU PLAN 5x & tube 50mm
https://farm5.staticflickr.com/4651/260 ... 84c5_o.jpg
BR, ADi
Does anybody know how to use the MTF mapper?
I have just tried with some test-pictures but the MTF-mapper has not recognized them up to now
Probably this program expects the whole test-picture. The quality of the whole test-picture scaled down is bad.
With the part of the test-picture it doesn’t work:
e.g. LSA & LU PLAN 5x & tube 50mm
https://farm5.staticflickr.com/4651/260 ... 84c5_o.jpg
BR, ADi
- rjlittlefield
- Site Admin
- Posts: 24018
- Joined: Tue Aug 01, 2006 8:34 am
- Location: Richland, Washington State, USA
- Contact:
ADi, somebody else may have other information, but as far as I can tell, MTF mapper is not going to be useful in your framework.
Quoting from https://sourceforge.net/projects/mtfmapper/ :
As seen in the image, the slanted edge is supposed to look black in the interior of the rectangle, white far outside the rectangle, and various gray across the transition, entirely as determined by the MTF of the camera system.
But of course as seen by a microscope objective, these laser-printed targets don't have anything resembling long straight edges. Instead they have varying densities of black dots with white highlights (specular reflections of the illumination), scattered at random across a light gray surface which is itself textured.
To use slanted edge method with a microscope objective, you would need a target that has long straight edges even as seen through the microscope objective. A carefully made chrome-on-glass target could conceivably be good enough, but even among the various test slides I have, including high-resolution USAF, I have never found an edge that I thought would be suitable for slant-edge testing.
I hope this depressing response will at least save you some wasted effort.
--Rik
Quoting from https://sourceforge.net/projects/mtfmapper/ :
The fundamental problem for you is that the slanted edge method requires targets that have long straight edges, white on one side and black on the other, with any irregularities not resolvable in the image.Automatically extracts dark (black) rectangular objects from images, and measures the Modulation Transfer Function (MTF, a measure of image sharpness) across the edges of the rectangles. Measurement is performed using the "slanted edge" method, similar to ISO 12233.
As seen in the image, the slanted edge is supposed to look black in the interior of the rectangle, white far outside the rectangle, and various gray across the transition, entirely as determined by the MTF of the camera system.
But of course as seen by a microscope objective, these laser-printed targets don't have anything resembling long straight edges. Instead they have varying densities of black dots with white highlights (specular reflections of the illumination), scattered at random across a light gray surface which is itself textured.
To use slanted edge method with a microscope objective, you would need a target that has long straight edges even as seen through the microscope objective. A carefully made chrome-on-glass target could conceivably be good enough, but even among the various test slides I have, including high-resolution USAF, I have never found an edge that I thought would be suitable for slant-edge testing.
I hope this depressing response will at least save you some wasted effort.
--Rik
Lou,Lou Jost wrote:I think it has to be a pure black vs white edge, not several edges of various shades of gray.
Wasn't that why folks used to burn the edge with a cigarette lighter, to darken the edge with black carbon?
Best,
Mike
Research is like a treasure hunt, you don't know where to look or what you'll find!
~Mike
~Mike
- rjlittlefield
- Site Admin
- Posts: 24018
- Joined: Tue Aug 01, 2006 8:34 am
- Location: Richland, Washington State, USA
- Contact:
Yes. And any graph like the one shown above should be an immediate clue that the something has gone badly wrong. What it shows is an MTF50 at 0.022 cycles per pixel, that is, over 45 pixels per cycle, around 23 pixels for a black/white transition. Further, the quite bizarre rise of MTF above 0.5 cycles per pixel, reaching almost MTF=2 at 1.0 cycles per pixel, must surely be some meaningless artifact.Lou Jost wrote:I think it has to be a pure black vs white edge, not several edges of various shades of gray.
Proper curves, indicating a likely good test, should look like the ones shown at http://mtfmapper.blogspot.com/2017/04/v ... n-gui.html , dropping smoothly and staying near zero to the right.
Razor blades are excellent targets at lower magnification, when properly blackened or back-lit so as to meet the black/white requirement. But their edges are too rough for use with a microscope objective, witness which we can clearly see the grinding marks even at the greatly reduced scale shown here.
--Rik
Edit: remove incorrect reference to Shannon sampling theorem, not relevant to any further posts.
Last edited by rjlittlefield on Mon Jan 29, 2018 5:45 pm, edited 1 time in total.