Filtering main light source vs post processing of RGB channels.

Here you can discuss topics such as focus stacking, stitching and other techniques that relate to the processing of micrographs.
Post Reply
Message
Author
Blue&green
Posts: 5
Joined: Fri Dec 13, 2019 9:22 pm

Filtering main light source vs post processing of RGB channels.

#1 Post by Blue&green » Sat Nov 14, 2020 12:23 am

Hi Everyone,

I would be interested to know peoples experimentation/opinion on the following specific to life science compound microscopy please;

In correcting light aberration using green spectrum exciting with Achromatic objectives,
The discernible results between;

Filtering light at source with a green filter vs using the manufacturers light source as intended and extracting the green layer from the RBG information post RAW Image?

Thank you in anticipation of your kind responses.

Regards Graham.

abednego1995
Posts: 103
Joined: Thu Dec 15, 2016 6:32 am

Re: Filtering main light source vs post processing of RGB channels.

#2 Post by abednego1995 » Sat Nov 14, 2020 3:21 am

Well, it would depend on how strict you want it to be. If it's just for getting rid of color and spherical aberrations, both would work OK. Post processing would be a bit harder since it's difficult to discern the best focus point in green while looking at an rgb image.

Another consideration would be, the spectral transmission of the filters. RGB filters on a color sensor have pretty broad transmission characteristics (and it's hard to find transmission curves for consumer sensors), whereas a dedicated interference filter would have tighter transmission properties with improved results.

Cheers,
John

Blue&green
Posts: 5
Joined: Fri Dec 13, 2019 9:22 pm

Re: Filtering main light source vs post processing of RGB channels.

#3 Post by Blue&green » Sat Nov 14, 2020 11:55 am

Thanks John,
I guess its going to be a trial between the two and see which is visually perceived sharper.

Chris Dee
Posts: 188
Joined: Mon Aug 12, 2019 1:02 pm

Re: Filtering main light source vs post processing of RGB channels.

#4 Post by Chris Dee » Sat Nov 14, 2020 1:40 pm

If your camera can shoot in RAW format the Camera Raw plugin for Photoshop has a 'remove chromatic aberration' feature. The colour range of the aberration you wish to remove and intensity can be set manually. Obviously only applicable for Photoshop users with a RAW capable camera, but worth mentioning as it works well.

Blue&green
Posts: 5
Joined: Fri Dec 13, 2019 9:22 pm

Re: Filtering main light source vs post processing of RGB channels.

#5 Post by Blue&green » Sat Nov 14, 2020 8:09 pm

Hi Chris, thank you for your reply.

Your comment was extremely useful, I have boycott Photoshop since the monthly fees because I can go three or four months without using it and considered it to be a waste of money keep paying out monthly when not using it so I paid and migrated to for Affinity Photo. I have discovered that Affinity has an equal command called "fringe" which does the same function. Interesting as before your comment I didn't know of this command.
I will have a look in a spare five minutes and see what I can learn.
Thank you again,

Regards Graham.

PS side note; Affinity Photo has a really good focus stack function which I have found incredibly good. I do not use Combine ZP now.

microb
Posts: 439
Joined: Wed Mar 27, 2019 6:39 am

Re: Filtering main light source vs post processing of RGB channels.

#6 Post by microb » Sat Nov 14, 2020 9:40 pm

Blue&green wrote:
Sat Nov 14, 2020 12:23 am
Filtering light at source with a green filter vs using the manufacturers light source as intended and extracting the green layer from the RBG information post RAW Image?
I don't know the images or light source, but camera sensitivity has green overlapping blue and red. So if the incoming data is bad, the green channel is going to capture that.

If you calibrate the blue and red responses, you could subtract their inputs from the green but that would probably over remove intensities leading to negative numbers, so you’ll have to then renormalize. It will still probably have good data removed though.
Attachments
d3afe2e69d5347368938e2c68e77b8a3.png
d3afe2e69d5347368938e2c68e77b8a3.png (101.22 KiB) Viewed 483 times

MichaelG.
Posts: 2435
Joined: Mon Apr 10, 2017 8:24 am
Location: NorthWest England

Re: Filtering main light source vs post processing of RGB channels.

#7 Post by MichaelG. » Sun Nov 15, 2020 10:01 am

microb wrote:
Sat Nov 14, 2020 9:40 pm
[…] camera sensitivity has green overlapping blue and red. So if the incoming data is bad, the green channel is going to capture that.

[…] It will still probably have good data removed though.
I think that is really the crux of the matter

Monochromatic lighting is fundamentally superior to post-processing

MichaelG.
Too many 'projects'

Chris Dee
Posts: 188
Joined: Mon Aug 12, 2019 1:02 pm

Re: Filtering main light source vs post processing of RGB channels.

#8 Post by Chris Dee » Sun Nov 15, 2020 10:36 am

MichaelG. wrote:
Sun Nov 15, 2020 10:01 am

Monochromatic lighting is fundamentally superior to post-processing

MichaelG.
Unless what you want to resolve emits the same wavelength as your monochromaticity?

Hobbyst46
Posts: 3018
Joined: Mon Aug 21, 2017 9:02 pm

Re: Filtering main light source vs post processing of RGB channels.

#9 Post by Hobbyst46 » Sun Nov 15, 2020 11:01 am

MichaelG. wrote:
Sun Nov 15, 2020 10:01 am
microb wrote:
Sat Nov 14, 2020 9:40 pm
[…] camera sensitivity has green overlapping blue and red. So if the incoming data is bad, the green channel is going to capture that.

[…] It will still probably have good data removed though.
I think that is really the crux of the matter

Monochromatic lighting is fundamentally superior to post-processing

MichaelG.
I believe this is true, if the software used to separate into R-G-B relies on sensor sensitivity.
The disadvantage of monochromatic light, though, is the relative low intensity if obtained from white light by means of a monochromator or a filter. A LED or laser is sufficiently bright though.
Zeiss Standard GFL+Canon EOS-M10, Olympus VMZ stereo

Post Reply