color blotches in stack

Starting out in microscopy? Post images and ask questions relating to the microscope and get answers from our more advanced users on the subject.

Moderators: rjlittlefield, ChrisR, Chris S., Pau

xys
Posts: 49
Joined: Thu Apr 21, 2011 11:42 am

color blotches in stack

Post by xys »

Is this is a Cyclops larva (= Nauplius) [as best as I can identify it]?

I did a 21 section focus stack using CombineZP. I get blue and green water-color-like splotches (I tried both Do Stack and Pyramoid Maximum Contrast, and get the color splotches from both routines). Am I doing something wrong or is this a normal result of stacking?

20x NA 0.65 very old (unplated brass barrel) B&L apochromat objective, 5x compensating (but not matched) projection lens.

I would appreciate critical comments or suggestions on any aspect of the image.

Many thanks.

Rashid

Image

rjlittlefield
Site Admin
Posts: 23564
Joined: Tue Aug 01, 2006 8:34 am
Location: Richland, Washington State, USA
Contact:

Re: color blotches in stack

Post by rjlittlefield »

xys wrote:I get blue and green water-color-like splotches (I tried both Do Stack and Pyramoid Maximum Contrast, and get the color splotches from both routines). Am I doing something wrong or is this a normal result of stacking?
The CombineZP Do Stack method simply selects pixel values from the original source frames, or from the source frames as modified by Align And Balance Used Frames if you ran that first. Splotches using that method can only happen if there is color at those places in one or another source frame. Pyramoid Maximum Contrast is more complicated, but again it's not prone to creating color from scratch.

Perhaps if you post out a couple of source frames too we can see what's going on.

It would also be easier to comment on the stack results in general, if we have a better idea what the source looks like.

From what I can see here, the stack looks pretty good except for a bit of what look like motion echoes on the front pair of legs.

How did you get the subject to hold still?

--Rik

xys
Posts: 49
Joined: Thu Apr 21, 2011 11:42 am

Post by xys »

Rik,

Many thanks for your quick reply and the detailed explanation. Just now was examining a picture that I took with the aperture closed for a large depth-of-field, and, you were right (as always!) it showed color splotches, so I was wrong to blame the stacking. I just logged in (too late) to correct my impression that this was a result of stacking. It is not as visible in the individual images at normal aperture, but I think it is there.

How did I get him to hold still? I didn't (either he was dead or he didn't have enough space to move between the slide and coverslip). I don't know if the eye being red indicates that he was alive.

I'll examine this color problem more carefully tomorrow and then will post a source frame as you suggested and also the frame with large depth of field.

Again, many thanks.

Rashid

xys
Posts: 49
Joined: Thu Apr 21, 2011 11:42 am

Post by xys »

Rik,

These are unprocessed (except for resizing and converting to sRGB). Left is one of the components used in the stack, Right is a small-aperture frame of the same object; it shows the color blotches more clearly than the one on the left (although they are there in both). I am wondering whether the colors are due to the ground glass covering the light source (this is the CHA Olympus student microscope in which the lamp is directly below a focusing lens whose lower surface is ground to diffuse the light, and directly below the condenser. I'll try to see if those blotches change as I alter the position of the condenser to focus on the ground glass (in the image plane) and away from the ground glass. Or it may be something else. I'll see if I can narrow this down, and will report back.

Thanks again.

Rashid

Image

rjlittlefield
Site Admin
Posts: 23564
Joined: Tue Aug 01, 2006 8:34 am
Location: Richland, Washington State, USA
Contact:

Post by rjlittlefield »

Thanks for the additional images. I don't know what all is going on here.

In the stacked image, there are pretty intense blue halos around some of the legs, that don't appear at all in these second images. I'm guessing those could be coming from some sort of chromatic aberration in the condenser, adding color to the edges of the legs when they're out of focus.

But there are also broad fuzzy patches of blue in the small-aperture single shot, far away from the subject in what should be uniformly colored background. I have no idea what those are about.

Sounds like it might be time to throw in a blank slide and see what that looks like.

--Rik

Pau
Site Admin
Posts: 6053
Joined: Wed Jan 20, 2010 8:57 am
Location: Valencia, Spain

Post by Pau »

xys wrote:...I am wondering whether the colors are due to the ground glass covering the light source (this is the CHA Olympus student microscope in which the lamp is directly below a focusing lens whose lower surface is ground to diffuse the light, and directly below the condenser. I'll try to see if those blotches change as I alter the position of the condenser to focus on the ground glass (in the image plane) and away from the ground glass...
I think you are right suspicing of some stuff in the light path. Also check for dirt on the illuminator surface. Moving the condenser up and down it will focus and defocus and change its color. Microscope condensers are not very well color corrected (even the expensive acromatic aplanatic ones). I'm dealing daily with a similar problem when I set my Zeiss Standard scope for Köler illumination because it has a flat glass just over the field diaphragm and any dirt on this glass shows up as blue or red phamtom images.
Combine ZP is not only stacking the images but also highly increasing its contrast, and both actions may be responsable of the increased visibility of this color patches. Try to turn off contrast in the satacking CZP macro (you can modify them)
Pau

xys
Posts: 49
Joined: Thu Apr 21, 2011 11:42 am

Post by xys »

Rik,

Many thanks for your further suggestions. This morning I discovered significant dirt on the upper surface of the polarizer. I put a blank slide as you suggested and I rotated the condenser while looking, but I could not see an effect in the image. But I took the condenser apart anyway and tried to clean all the surfaces. Then using each of the dry objectives in turn, I took three images, one with the condenser (a) all the way up, (b) adjusted to show the ground glass above the light source in the image plane, and (c) a position in between a and b. Pau must have communicated this to me by telepathy, as I read his suggestion after I had done this :)

Image

The above images of the blank slide were taken with the 10x objective:

A. Condenser as high as it will go. Post-processing: levels, then "remove color cast" in Elements. There is barely a hint of color but the background is blotchy (perhaps because I used levels on a blank slide which would greatly exaggerate uneven lighting).

B. Condenser somewhat higher than position that will show the groundglass in focus, but somewhat lower than I normally use it at. Post processing: levels only. Mottled brownish with hints of blue or green.

C: Same frame as B, but after the levels adjustment, I did "remove color cast". This looks like an extreme case of what the Cyclops picture showed.

My tentative and probably wrong conclusion: the ground glass forming the upper part of the illuminator is acting like many small irregular prisms refracting the light and resulting in the "colorful" image, made worse (more vivid) when color is adjusted to gray background in post processing. This is probably due to my ignorance. I probably am misadjusting the condenser position, as in the absence of Koehler illumination, I really don't know where the condenser should be. I may simply need to rack the condenser all the way up to get a somewhat cleaner background (as in image A).

Of course, I would be most grateful for your suggestions (even as I feel I've already taxed everyone's patience).

Many thanks,

Rashid

xys
Posts: 49
Joined: Thu Apr 21, 2011 11:42 am

Post by xys »

Pau,

Many thanks again for all your help. As I mentioned above in my reply to Rik, I got your message telepathically and did exactly (I think) what you had kindly suggested. I think this is all due to my not knowing how to adjust this microscope since it lacks the field diaphragm in microscopes with Koehler illuminators. In my previouse post in reply to Rik, I have examples of the images of a blank slide at two positions of the condenser.

Many thanks again for your comments and suggestions.

Rashid

rjlittlefield
Site Admin
Posts: 23564
Joined: Tue Aug 01, 2006 8:34 am
Location: Richland, Washington State, USA
Contact:

Post by rjlittlefield »

xys wrote:I may simply need to rack the condenser all the way up to get a somewhat cleaner background (as in image A).
I think this would be a good approach.

The essence of Köhler illumination is to be sure that the light source is completely defocused in the image plane, by instead focusing it at the back of the objective. When there is a diffusion element in the light path, then the closest equivalent of Köhler illumination would be to focus the diffusion element at the back of the objective. In that case there would be no trace of the diffusion element in the specimen plane. It's quite possible that your condenser cannot be adjusted to do that, so you'll have to pick whatever compromise works best by experiment.

One thing you can definitely do to improve your situation is to set color balance on your camera so that a blank slide is captured as neutral gray or close to it. Making big changes in color balance in post-processing is liable to mess up whatever color shifts there are in the actual image, both blotches from the illumination and real color in the specimen.

--Rik

xys
Posts: 49
Joined: Thu Apr 21, 2011 11:42 am

Post by xys »

Rik,

Thank you once again. I'll do what you suggest. And yes, I simply didn't think to take a custom white balance. I usually aim to do so, but even then, I sometimes I forget and change the lamp voltage losing the white balance.

Rashid

Post Reply Previous topicNext topic