MicrobeHunter.com Microscopy Forum

Hello! I just started taking pictures yesterday with an Omax A35140U camera. Yesterday, I was pretty happy with it. Today, I fired it up, and it has horrible color distortion, green towards the center and red towards the corners.

Here's an imgur album showing the difference: The first picture is from yesterday, and the second is from today: https://imgur.com/a/QjGkI9s

This distortion seems to be happening no matter what light level, condenser iris width, condenser height, etc. I use. I'm using ToupView, which has a white balance option, but it doesn't help - balancing toward the middle just makes the corners redder, and balancing toward the corners just makes the center greener.

Can anyone provide any suggestions?

Maybe a stupid tip but have you removed the protective plastic for the sensor that is supposed to protect against dust? //KD

I don't believe there was one. I don't see one on it.

Also, whatever is happening now wasn't happening yesterday.

Hard to judge in the first image, but in the second the greenish/purplish fringing is centered way off toward the lower right corner of the image. It is possible the camera tube got bumped out of alignment? Also to clarify, you see this in the camera only, not through the eyepieces, right?

Yes, sorry - I meant to say, this is only in the camera. The eyepieces show pure white background.

I suppose it's possible, but I've tried re-seating the camera, and the colors don't change.

Edit: Also, I probably wasn't clear - the first image is to show how it used to be. I'd like to get it back to that quality.

If you center an object in the field of view looking through the eyepieces is it also reasonably close to centered in the camera view?

If you set up a similar sample to that in the first image ; a relatively large structure and only about 50% background, does the camera then behave itself? It looks like it is a metering issue and the minimal area, scattered sample in your second example is not providing a defined enough area for the meter to fix on and adjust the exposure and balance correctly. Is there a setting for the type of light?

It's pretty close to centered, just a little bit off to one side. In this one, I focused the small rectangular crystal to the center of the eyepieces, and this is how the camera saw it: https://imgur.com/a/GR7IGWd

Here's a comparable image to the first example image, in terms of field covered by the sample versus background. It still has the same color distortion: https://imgur.com/a/ZURrff3

For reference, here are some more pictures I took on the first day. None of them have that color distortion: https://imgur.com/a/Fa8bN00

Edit: I should point out, the two in polarized light do look like they have a bit of red distortion in the corner, but that's actually an artifact of the polarization - I couldn't get the background quite even because I was rotating rectangular filters by hand. In the brightfield pics, you can see it's not there.

Still not obvious to me whether this is an optical or software issue. A photo of your microscope with camera may help. It looks like that model camera fits into a normal eyepiece tube? Is it possible to swap the camera into the binocular and one of the normal eyepieces into the camera port and see if the results are the same? I looked at the histogram of the image with the small rectangular crystal in the center and saw no obvious software-induced artifacts:

That was an excellent idea, though the results are not encouraging. An eyepiece does indeed fit into the photo tube, and does not show distortion. The camera in the eyepiece does show it.

Are the camera and eyepiece views in focus at the same time or do you have to refocus between them? It looks like the camera consists of a C-mount relay lens, C-to-CS mount spacer, and CS-mount camera, any chance those threaded joints are loose? You could try unscrewing and removing the lens from the camera and see if the same background colors still show up with unfocused/diffuse ambient light hitting the sensor.

I think you may be on to something - when I let diffuse light hit the camera without the relay lens or the spacer, I get an even field of color. If I let diffuse light hit it with the relay lens or spacer, it's a little difficult to tell, but I think the distortion may be there. It's darker toward the corners, at least.

Edit: to answer your question, yes, I have to refocus between the eyepieces and the camera tube a little.

So probably not a hardware/software issue with the camera itself, what about the question of whether the camera and eyepiece views are in focus at the same time?

Just saw your edit regarding focus. If you focus the camera then without turning the focus knobs lift an eyepiece until it is also in focus (or vice versa, depending on the direction of mismatch between the two) how far do you have to lift to get them in focus at the same time?

Not very far at all - maybe three quarters of a centimeter.

I think 3/4 of a centimeter is a pretty big mismatch, worth investigating. Is the focus mismatch between the binocular and camera port the same with an eyepiece in the camera port? Or only when using the camera?

I have seen similar problems on a much lower resolution eyepiece camera

For no immediately obvious reason, I suspect the 0.5x reduction lens : They seem to be very sensitive to alignment [possibly related to the tiny 1.4 micron pixels].

Can you rig the camera directly [C-mount] without the reduction lens, on your ‘scope ? ... it would be a useful check, I think.

MichaelG.

.
With apologies for drifting off-topic ... I’ve just found this page from Meiji, which might be of interest to readers: https://meijitechno.com/camera-mounts/

Hi, the latest Toupview is able to fix background on the fly (i.e. at capture) - use the 'flat field correction' - just capture blank area of slide first (as the dropdown-box will show you) then select to use it - it's amazingly good! Good tip for Toupview - NEVER use sharpening in live-view - it cripples the frame-rate and makes you think your camera's faulty!

Holy cow, that's just what I'm looking for! Thanks, mrsonchus. My mic's dust-covered and put away for the night, but I'll try that tomorrow.

For what it's worth, a few more tests definitely seem to have narrowed the problem down to the reduction lens. Is there some place to get a replacement for that? Omax doesn't seem to sell them separately (or at least, I can't find them)

I use this focusabale (great to set parfocality with eyepieces) version, from Brunel Microscopes or of course elsewhere - they're all ToupCams as far as I know.
See my post
See also at Touptec

It is interesting to see that [unless I have missed something]:

Omax does not list the better reduction lens, and
Touptek does not list the basic one

I also note, from the product page, a hint of spurious background colour in the sample images !
https://omaxmicroscope.com/microscope-a ... linux.html

The jury is still out, but I’m inclined to think that the basic 0.5x reduction lens is ‘guilty as charged’

MichaelG.

MichaelG. wrote: ↑

Sun Aug 02, 2020 10:31 am

I also note, from the product page, a hint of spurious background colour in the sample images !
https://omaxmicroscope.com/microscope-a ... linux.html

If you refer to the image of some bluish-stained plant cells (and 2 other images), with a yellowish background on the right half of the image, "a hint of spurious" is a very mild description...

I thought that these 0.5X reduction lenses, that are bundled (on eBay at least) with eyepiece cameras and other c-mount cameras, are really a mystery with respect to construction (how many elements? achromats ? any corrections ?). What would the jury say when they return and face the judge ?

Hobbyst46 wrote: ↑

Sun Aug 02, 2020 10:54 am

If you refer to the image of some bluish-stained plant cells (and 2 other images), with a yellowish background on the right half of the image, "a hint of spurious" is a very mild description...

English Gentlemen are renowned for understatement

MichaelG.

Jerradin wrote: ↑

Sun Aug 02, 2020 5:54 am

For what it's worth, a few more tests definitely seem to have narrowed the problem down to the reduction lens. Is there some place to get a replacement for that?

Based on your photos and the 3/4 centimeter focus mismatch between the binocular and camera port (assuming you have verified that the mismatch happens only with the camera and not with an eyepiece in the camera port) I still suspect there may be nothing wrong with the lens itself, rather the lens is simply not positioned the correct distance from the camera sensor. Referring to this Omax product page:

1. Is the CS-to-C mount spacer ring marked "5mm" present in your setup?
2. It looks like the camera itself has a focus adjustment built into the body: if you loosen a set screw on the side, the entire ring that the lens threads into can itself be threaded in/out of the camera body. (I have different CS mount camera with the same sort of adjustment.) Is it possible the set screw is loose and the camera's focus adjustment has been changing as you remove/install the lens?

I don't see the spacer anywhere in my setup. The ring that the lens is in seems quite sturdy and in-place.

I'm running into something strange with the flat field correction - It works perfectly in the video preview, but when I hit "snap" to take a high-res picture, the picture has the same old color distortion. Does it not apply to stills? I do have "enable" checked.

Jerradin wrote: ↑

Sun Aug 02, 2020 11:12 pm

I don't see the spacer anywhere in my setup.

Try focusing through the eyepieces, then without changing the focus on the microscope, unscrew the lens until the camera is in focus. (Assuming that is the direction of mismatch, and the lens doesn't fall out of the camera before you get focus...)

Another thought occurs - this is very similar to the problems I had with a beautiful Orthoplan - the cause was the impossibility (for me at least) of perfect alignment of the lamp - I could never get rid of this effect and did try everything - it was the alignment of the illumination. What type of microscope do you have and is it possible to check, or even just remove then re-fit the lamp and/or the head? Either of these may well have been fine first time then have moved subsequently....

Perhaps remove the bulb then put it back in, and/or unscrew the head's clamping screw, take it off, then put it back on...

I'd try the bulb first, on it's own.

I'm pretty sure it's not anything to do with the scope itself, given that the camera produces the same color distortions when it's removed from the microscope along with the reduction lens, and an eyepiece in the photo tube doesn't see the distortion.

Thanks for the help everyone, I think I've got it narrowed down to A. get a better reduction lens/spacer setup at some point, and B. use ToupView's field correction in the meantime. It's not a perfect solution because the color variance seems to increase depending on how much light is coming through the lens, so when the light levels change the field correction stops being as effective, but it's workable for now.

Oh, also, I wanted to say that I've fixed the difference in focal points between the binoculars and the photo tube, too. In the microscope setup manual, it says to set the eyepiece diopter values to the same as the interpupillary distance, then set the focus from there. That seems to be what causes the difference, though - if I begin the focusing with the diopter values closer to 0, the difference goes away.

Ah yes, with TV's correction (as with other manual techniques of using a pre-shot plain background sample) it's important to keep the exposure and objective the same for each background sample taken. Tv takes a background when you select the option and set how many 'takes' to use... What this practically means is that exposure must not be set to auto - and objective changes will need a new background set. Easy enough as TV does it so quickly - if you change objectives simply re-calculate the background with the TV facility.
In fact I've been using it (the correction method) today with my 5mp Toupcam atop the trinoc of my stereo-zoom 'scope - no problems.

I contacted Omax's customer support to see what ideas they had, and the rep's response indicated that he couldn't tell what I meant by color distortion. This is the album I sent them: https://imgur.com/a/UkICDDV

I guess I need a bit of a sanity check? It seems pretty dang obvious to me.