LED Lighting and optical resolution questions
-
- Posts: 1167
- Joined: Tue Jul 07, 2020 1:51 pm
- Location: Scotland
Re: LED Lighting and optical resolution questions
I did some comparisons on the Swift between two cameras - the Canon 1100d and a ToupTek 5MP eyepiece camera. Imaged the captive diatom at 40x and did some on-screen measurements at 100% crop:
Canon 1100d Touptek: Both gave a length of 63um for the diatom, which was reassuring The Touptek gave a bigger image by a factor of 2.36 which is the ratio of the pixel sizes - 5.19/2.2 = 2.36. The striae on the bigger Touptek were slightly clearer. and a measurement of the average distance between striae gave a value of 0.91um (if my calcs were correct!). I've no doubt a properly prepared diatom would give clearer striae. A guesstimate of the potential resolution using the Touptek with the 40x objective was maybe 2/3 of that i.e. ~600nm (but open to discussion!).
Anyone happen to recognise the diatom species? A type of Frustulia?
Anyway, I think I'd conclude that there may well be an advantage to using a camera with smaller pixels than the Canon1100d.
Thanks
Louise
Canon 1100d Touptek: Both gave a length of 63um for the diatom, which was reassuring The Touptek gave a bigger image by a factor of 2.36 which is the ratio of the pixel sizes - 5.19/2.2 = 2.36. The striae on the bigger Touptek were slightly clearer. and a measurement of the average distance between striae gave a value of 0.91um (if my calcs were correct!). I've no doubt a properly prepared diatom would give clearer striae. A guesstimate of the potential resolution using the Touptek with the 40x objective was maybe 2/3 of that i.e. ~600nm (but open to discussion!).
Anyone happen to recognise the diatom species? A type of Frustulia?
Anyway, I think I'd conclude that there may well be an advantage to using a camera with smaller pixels than the Canon1100d.
Thanks
Louise
A Nikon CF plan 20x; A Swift 380T; A DIY infinity corrected focus rail system with a 40x/0.65 Olympus Plan, a 10x/0.30 Amscope Plan Fluor, and a 20x/0.75 Nikon Plan Apo
Re: LED Lighting and optical resolution questions
A point which has been previously discussed [sometimes with a little difficulty] on this forum.LouiseScot wrote: ↑Mon Nov 16, 2020 2:11 pmAnyway, I think I'd conclude that there may well be an advantage to using a camera with smaller pixels than the Canon1100d.
Keep at it, Louise ... Your results will be a benchmark in future debates I’m sure.
MichaelG.
Too many 'projects'
Re: LED Lighting and optical resolution questions
It looks like the ToupTek software may be boosting contrast quite a bit more than Canon? Also looks like the Canon image may be suffering more from noise and subsequent noise reduction? What ISO setting are you using on the Canon? I do not get the impression that pixel size is currently the limiting factor in the Canon image.
-
- Posts: 1167
- Joined: Tue Jul 07, 2020 1:51 pm
- Location: Scotland
Re: LED Lighting and optical resolution questions
I think that's probably just my post processing. Having said that I'm wondering whether I'm doing the auto exposure control in the Canon right. I'll try using a centre weighted evaluation next time. The Toupview software also does auto exposure but it can seemingly take ages for it to work it out (I'm impatient...). Also the camera only allows certain exposure times. On the other hand the Toupview does allow variable gain. I'm thinking about either getting a better dedicated camera for the Swift or maybe an actual dslr adapter e.g. https://www.aliexpress.com/item/3268924 ... 2257eJuJOv. Neither of which helps me improve the images I'm getting from the rail system so I'll have to keep working on that! I think I'll try one of my mono astro cameras out, just to compare some more. I have a 6MP one which should give good images.hans wrote: ↑Mon Nov 16, 2020 7:33 pmIt looks like the ToupTek software may be boosting contrast quite a bit more than Canon? Also looks like the Canon image may be suffering more from noise and subsequent noise reduction? What ISO setting are you using on the Canon? I do not get the impression that pixel size is currently the limiting factor in the Canon image.
Thanks
Louise
A Nikon CF plan 20x; A Swift 380T; A DIY infinity corrected focus rail system with a 40x/0.65 Olympus Plan, a 10x/0.30 Amscope Plan Fluor, and a 20x/0.75 Nikon Plan Apo
Re: LED Lighting and optical resolution questions
Since astro cameras, per definition maybe, are supposedly highly light-sensitive, relative to common inexpensive microscope eyepiece cameras, the results should be very interesting.LouiseScot wrote: ↑Mon Nov 16, 2020 8:11 pm...I think I'll try one of my mono astro cameras out, just to compare some more. I have a 6MP one which should give good images.
-
- Posts: 1167
- Joined: Tue Jul 07, 2020 1:51 pm
- Location: Scotland
Re: LED Lighting and optical resolution questions
I'll let you know! I'm going to have a go today I'll probably have to dial down the LED brightness a bit. I'll try it with blue and green narrowband filters. I've been looking at Arturo Agostino's DIC images - I can't compete with them!Hobbyst46 wrote: ↑Tue Nov 17, 2020 12:43 pmSince astro cameras, per definition maybe, are supposedly highly light-sensitive, relative to common inexpensive microscope eyepiece cameras, the results should be very interesting.LouiseScot wrote: ↑Mon Nov 16, 2020 8:11 pm...I think I'll try one of my mono astro cameras out, just to compare some more. I have a 6MP one which should give good images.
I've just ordered a 2x dslr adapter off Ebay. It probably won't improve my picture quality but should overcome the vignetting on the Swift. Fingers crossed I don't get stung for vat/import duty... ha ha.
Louise
A Nikon CF plan 20x; A Swift 380T; A DIY infinity corrected focus rail system with a 40x/0.65 Olympus Plan, a 10x/0.30 Amscope Plan Fluor, and a 20x/0.75 Nikon Plan Apo
-
- Posts: 1167
- Joined: Tue Jul 07, 2020 1:51 pm
- Location: Scotland
Re: LED Lighting and optical resolution questions
So here are some images taken with the ToupTek cam on the Swift 380T, then with the mono astro cam (Sony 178m sensor), and then with the astro cam + a 486nm narrowband filter. I couldn't find the previous diatom I imaged - I know it's there somewhere! Anyway, found this other one. I should mention I've been using a cavity slide plus cover slip - not sure if a cavity slide has a negative effect on image quality?
Touptek cam 40x:
at 50%:
Astro cam 40x:
at 50%:
Astro cam narrowband 486nm:
There you go
I suppose these are about as good as I'm going to get with the Swift though I'd like to think I could somehow get better images. Apart from the colour, obviously, no big difference between the Touptek and the Astro cam, as far as I can see. I think cropping images at 100% size is maybe not such a good idea - they always appear to come out blurry whatever camera or objective I use. I did a calculation for the images I posted yesterday (with the scales). The diatom was 63um in length, the length on the screen of the Touptek image was 146mm so that's an enlargement of 146/.063 = 2317x which seems rather a lot! The actual magnification was 40x and it does seem tiny looking at it via the eyepieces!
Louise
Touptek cam 40x:
at 50%:
Astro cam 40x:
at 50%:
Astro cam narrowband 486nm:
There you go
I suppose these are about as good as I'm going to get with the Swift though I'd like to think I could somehow get better images. Apart from the colour, obviously, no big difference between the Touptek and the Astro cam, as far as I can see. I think cropping images at 100% size is maybe not such a good idea - they always appear to come out blurry whatever camera or objective I use. I did a calculation for the images I posted yesterday (with the scales). The diatom was 63um in length, the length on the screen of the Touptek image was 146mm so that's an enlargement of 146/.063 = 2317x which seems rather a lot! The actual magnification was 40x and it does seem tiny looking at it via the eyepieces!
Louise
A Nikon CF plan 20x; A Swift 380T; A DIY infinity corrected focus rail system with a 40x/0.65 Olympus Plan, a 10x/0.30 Amscope Plan Fluor, and a 20x/0.75 Nikon Plan Apo
Re: LED Lighting and optical resolution questions
In this case I think that the astro camera with or without filter performed less sharply than the other camera. When I convert the color photo of the diatom to grayscale, it is less blurred than the astro camera images.
Live diatoms usually move in water. I would suspect that is the chief reason of the blurry. The cavity slide only promotes swimming
Live diatoms usually move in water. I would suspect that is the chief reason of the blurry. The cavity slide only promotes swimming
-
- Posts: 762
- Joined: Thu Aug 23, 2018 10:12 pm
- Location: Lund, Sweden
Re: LED Lighting and optical resolution questions
I would suggest you redo your tests with regular slides, and try to wick away as much water as possible (or use prepared slides). A thick layer of water between the cover glass and the subject will deteriorate the image substantially, and the effect will be stronger for high NA objectives (like the 20x 0.75) than lower NA ones.
I'm personally pretty bad at sample prep, I always put too much gunk on the slide, so I always use my Klaus Kemp 8 form slide when I want to perform standardized tests.
I'm personally pretty bad at sample prep, I always put too much gunk on the slide, so I always use my Klaus Kemp 8 form slide when I want to perform standardized tests.
Re: LED Lighting and optical resolution questions
Very much so! You want a flat slide, a cover slip that is 0.17mm or very near to that and minimum water between the diatom and the bottom of the cover slip.LouiseScot wrote: ↑Tue Nov 17, 2020 4:17 pmI should mention I've been using a cavity slide plus cover slip - not sure if a cavity slide has a negative effect on image quality?
Ideally, as mentioned, you want a Test Slide ... bite the bullet
Zeiss Standard WL (somewhat fashion challenged) & Wild M8
Olympus E-P2 (Micro Four Thirds Camera)
Olympus E-P2 (Micro Four Thirds Camera)
-
- Posts: 1167
- Joined: Tue Jul 07, 2020 1:51 pm
- Location: Scotland
Re: LED Lighting and optical resolution questions
Although there is water still there all the diatoms are dead - they are ex diatoms! I've messed up that particular slide now so will have to make up a fresh one and with a flat slide. I find the diatoms very difficult to focus on with the Swift. Since I had to change focus for the different cameras it may just be a difference in focus rather than quality.Hobbyst46 wrote: ↑Tue Nov 17, 2020 4:49 pmIn this case I think that the astro camera with or without filter performed less sharply than the other camera. When I convert the color photo of the diatom to grayscale, it is less blurred than the astro camera images.
Live diatoms usually move in water. I would suspect that is the chief reason of the blurry. The cavity slide only promotes swimming
Thanks
Louise
A Nikon CF plan 20x; A Swift 380T; A DIY infinity corrected focus rail system with a 40x/0.65 Olympus Plan, a 10x/0.30 Amscope Plan Fluor, and a 20x/0.75 Nikon Plan Apo
-
- Posts: 1167
- Joined: Tue Jul 07, 2020 1:51 pm
- Location: Scotland
Re: LED Lighting and optical resolution questions
Yeah, I thought that might be a problem so I'll use flat slides for any other goes. I'm only using the Swift for now. I looked up Klaus Kemp - he's not taking any new orders at the moment. I only have the 3 vintage slides with max NA of 0.65 for the 40x achro. There are other diatoms in my culture on the window sillviktor j nilsson wrote: ↑Tue Nov 17, 2020 5:18 pmI would suggest you redo your tests with regular slides, and try to wick away as much water as possible (or use prepared slides). A thick layer of water between the cover glass and the subject will deteriorate the image substantially, and the effect will be stronger for high NA objectives (like the 20x 0.75) than lower NA ones.
I'm personally pretty bad at sample prep, I always put too much gunk on the slide, so I always use my Klaus Kemp 8 form slide when I want to perform standardized tests.
Thanks
Louise
A Nikon CF plan 20x; A Swift 380T; A DIY infinity corrected focus rail system with a 40x/0.65 Olympus Plan, a 10x/0.30 Amscope Plan Fluor, and a 20x/0.75 Nikon Plan Apo
-
- Posts: 1167
- Joined: Tue Jul 07, 2020 1:51 pm
- Location: Scotland
Re: LED Lighting and optical resolution questions
Ok. As mentioned I have the vintage test slides. I'll have a look to see if I can find some contemporary ones but, if there are any, they are probably rather expensive....75RR wrote: ↑Tue Nov 17, 2020 5:51 pmVery much so! You want a flat slide, a cover slip that is 0.17mm or very near to that and minimum water between the diatom and the bottom of the cover slip.LouiseScot wrote: ↑Tue Nov 17, 2020 4:17 pmI should mention I've been using a cavity slide plus cover slip - not sure if a cavity slide has a negative effect on image quality?
Ideally, as mentioned, you want a Test Slide ... bite the bullet
Louise
A Nikon CF plan 20x; A Swift 380T; A DIY infinity corrected focus rail system with a 40x/0.65 Olympus Plan, a 10x/0.30 Amscope Plan Fluor, and a 20x/0.75 Nikon Plan Apo
Re: LED Lighting and optical resolution questions
PM sent.LouiseScot wrote: ↑Tue Nov 17, 2020 6:40 pmOk. As mentioned I have the vintage test slides. I'll have a look to see if I can find some contemporary ones but, if there are any, they are probably rather expensive....75RR wrote: ↑Tue Nov 17, 2020 5:51 pmVery much so! You want a flat slide, a cover slip that is 0.17mm or very near to that and minimum water between the diatom and the bottom of the cover slip.LouiseScot wrote: ↑Tue Nov 17, 2020 4:17 pmI should mention I've been using a cavity slide plus cover slip - not sure if a cavity slide has a negative effect on image quality?
Ideally, as mentioned, you want a Test Slide ... bite the bullet
Louise
-
- Posts: 1167
- Joined: Tue Jul 07, 2020 1:51 pm
- Location: Scotland
Re: LED Lighting and optical resolution questions
I've just had a quick try on the Swift of using oblique illumination. With the Pleurosigma vintage slide at 40x and Touptek camera.
50% crop: 100% crop: The little adjacent diatom is rather nice. The oblique lighting certainly increases contrast! I'll have to look further into how best to use it.
Louise
50% crop: 100% crop: The little adjacent diatom is rather nice. The oblique lighting certainly increases contrast! I'll have to look further into how best to use it.
Louise
A Nikon CF plan 20x; A Swift 380T; A DIY infinity corrected focus rail system with a 40x/0.65 Olympus Plan, a 10x/0.30 Amscope Plan Fluor, and a 20x/0.75 Nikon Plan Apo
Re: LED Lighting and optical resolution questions
Nice!
My take is that Brightfield should only be used as default illumination i.e. the first step in setting up the microscope.
This includes finding and focusing on the specimen, adjusting initial lamp brightness and setting up Köhler.
At that point one should then bring in one of the main illumination methods such as Darkfield, Oblique, COL, Phase, DIC etc
The most rewarding cost wise is of course Oblique, the only downside is the lack of consistency/repeatability in its effects, as it depends greatly on the skill of the microscopist.
My take is that Brightfield should only be used as default illumination i.e. the first step in setting up the microscope.
This includes finding and focusing on the specimen, adjusting initial lamp brightness and setting up Köhler.
At that point one should then bring in one of the main illumination methods such as Darkfield, Oblique, COL, Phase, DIC etc
The most rewarding cost wise is of course Oblique, the only downside is the lack of consistency/repeatability in its effects, as it depends greatly on the skill of the microscopist.
Zeiss Standard WL (somewhat fashion challenged) & Wild M8
Olympus E-P2 (Micro Four Thirds Camera)
Olympus E-P2 (Micro Four Thirds Camera)
-
- Posts: 1167
- Joined: Tue Jul 07, 2020 1:51 pm
- Location: Scotland
Re: LED Lighting and optical resolution questions
I wish someone had mentioned that to me before! But, never mind - live and learn! I'm going to try and 3d print a COL filter today75RR wrote: ↑Thu Nov 19, 2020 7:26 amNice!
My take is that Brightfield should only be used as default illumination i.e. the first step in setting up the microscope.
This includes finding and focusing on the specimen, adjusting initial lamp brightness and setting up Köhler.
At that point one should then bring in one of the main illumination methods such as Darkfield, Oblique, COL, Phase, DIC etc
The most rewarding cost wise is of course Oblique, the only downside is the lack of consistency/repeatability in its effects, as it depends greatly on the skill of the microscopist.
Louise
A Nikon CF plan 20x; A Swift 380T; A DIY infinity corrected focus rail system with a 40x/0.65 Olympus Plan, a 10x/0.30 Amscope Plan Fluor, and a 20x/0.75 Nikon Plan Apo
-
- Posts: 762
- Joined: Thu Aug 23, 2018 10:12 pm
- Location: Lund, Sweden
Re: LED Lighting and optical resolution questions
Agree, looking good! Quite a lot of chromatic aberration, but that's expected with basic achromats. Should be easy to improve with some post processing. But sharpness and resolution looks good. Nicely illuminated, good use of oblique. Will be interesting to see a comparison with your better objectives, now that you've found a good test slide to use.
-
- Posts: 1167
- Joined: Tue Jul 07, 2020 1:51 pm
- Location: Scotland
Re: LED Lighting and optical resolution questions
I'm not worried about small amounts of CA. I shall continue with my investigative experiments Oh, it's the same Pleurosigma vintage slide I used before.viktor j nilsson wrote: ↑Thu Nov 19, 2020 9:04 amAgree, looking good! Quite a lot of chromatic aberration, but that's expected with basic achromats. Should be easy to improve with some post processing. But sharpness and resolution looks good. Nicely illuminated, good use of oblique. Will be interesting to see a comparison with your better objectives, now that you've found a good test slide to use.
Louise
A Nikon CF plan 20x; A Swift 380T; A DIY infinity corrected focus rail system with a 40x/0.65 Olympus Plan, a 10x/0.30 Amscope Plan Fluor, and a 20x/0.75 Nikon Plan Apo
-
- Posts: 762
- Joined: Thu Aug 23, 2018 10:12 pm
- Location: Lund, Sweden
Re: LED Lighting and optical resolution questions
Oh, I should've noticed that.LouiseScot wrote: Oh, it's the same Pleurosigma vintage slide I used before.
Louise
-
- Posts: 1167
- Joined: Tue Jul 07, 2020 1:51 pm
- Location: Scotland
Re: LED Lighting and optical resolution questions
There's quite a few Pleurosigma on that slide, plus a few other species, so each time I use it I'll get a different viewviktor j nilsson wrote: ↑Thu Nov 19, 2020 10:14 amOh, I should've noticed that.LouiseScot wrote: Oh, it's the same Pleurosigma vintage slide I used before.
Louise
Louise
A Nikon CF plan 20x; A Swift 380T; A DIY infinity corrected focus rail system with a 40x/0.65 Olympus Plan, a 10x/0.30 Amscope Plan Fluor, and a 20x/0.75 Nikon Plan Apo
Re: LED Lighting and optical resolution questions
I like these images.viktor j nilsson wrote: ↑Thu Nov 19, 2020 9:04 amAgree, looking good! Quite a lot of chromatic aberration, but that's expected with basic achromats. Should be easy to improve with some post processing. But sharpness and resolution looks good. Nicely illuminated, good use of oblique. Will be interesting to see a comparison with your better objectives, now that you've found a good test slide to use.
I think that with diatoms, light refraction, and even reflections, in the glass-like frustule is more significant than the type of objective. I get the same chromatic effects on diatoms with Neofluars. Never gave it a serious attention, though.
-
- Posts: 1167
- Joined: Tue Jul 07, 2020 1:51 pm
- Location: Scotland
Re: LED Lighting and optical resolution questions
A quick go with COL. I think the colour fringing is a consequence of the oblique technique rather than the property of the lens per se.viktor j nilsson wrote: ↑Thu Nov 19, 2020 9:04 amAgree, looking good! Quite a lot of chromatic aberration, but that's expected with basic achromats. Should be easy to improve with some post processing. But sharpness and resolution looks good. Nicely illuminated, good use of oblique. Will be interesting to see a comparison with your better objectives, now that you've found a good test slide to use.
40x again, but limited to 50% of the image size. Seems to be better than 100%.
First one, focus is biased towards the top of the Pleurosigma:
I find it difficult to get a sharp focus with the Swift. Of course, I'll also try this out on the rail system with the better lenses.
Louise
- Attachments
-
- Pleurosigma40x_col2_sideadj50pc.jpg (98.79 KiB) Viewed 9355 times
Last edited by LouiseScot on Thu Nov 19, 2020 12:22 pm, edited 1 time in total.
A Nikon CF plan 20x; A Swift 380T; A DIY infinity corrected focus rail system with a 40x/0.65 Olympus Plan, a 10x/0.30 Amscope Plan Fluor, and a 20x/0.75 Nikon Plan Apo
Re: LED Lighting and optical resolution questions
It would be interesting to see what that entailed and the source you got it from.
By the way, your images labeled Pleurosigma40x_col2_top50pc.jpg and Pleurosigma40x_col2_sideadj50pc.jpg look the same to me. Mislabeled?
Zeiss Standard WL (somewhat fashion challenged) & Wild M8
Olympus E-P2 (Micro Four Thirds Camera)
Olympus E-P2 (Micro Four Thirds Camera)
-
- Posts: 1167
- Joined: Tue Jul 07, 2020 1:51 pm
- Location: Scotland
Re: LED Lighting and optical resolution questions
Yeah, I accidentally saved one on top of the other, if you know what I mean! D'uh. Have now deleted the second one...
Actually, in the end for the COL I just used an iris diaphragm plus a an opaque central stop (as used for dark field) which I'd previously 3D printed. It might be better with a slightly smaller stop than I used. I'll have to experiment to see what works best. There is some info about COL here:
http://www.microscopy-uk.org.uk/mag/artdec02/pjcol.html
Louise
A Nikon CF plan 20x; A Swift 380T; A DIY infinity corrected focus rail system with a 40x/0.65 Olympus Plan, a 10x/0.30 Amscope Plan Fluor, and a 20x/0.75 Nikon Plan Apo
Re: LED Lighting and optical resolution questions
Paul James is my go to person on COL as wellLouiseScot wrote: ↑Thu Nov 19, 2020 12:32 pmActually, in the end for the COL I just used an iris diaphragm plus a an opaque central stop (as used for dark field) which I'd previously 3D printed. It might be better with a slightly smaller stop than I used. I'll have to experiment to see what works best. There is some info about COL here:
http://www.microscopy-uk.org.uk/mag/artdec02/pjcol.html
Don't know if this will be usefull to you, they increase in size by 1mm
download/file.php?id=335
Zeiss Standard WL (somewhat fashion challenged) & Wild M8
Olympus E-P2 (Micro Four Thirds Camera)
Olympus E-P2 (Micro Four Thirds Camera)
-
- Posts: 1167
- Joined: Tue Jul 07, 2020 1:51 pm
- Location: Scotland
Re: LED Lighting and optical resolution questions
Thanks - though they are simple to 3D print and I already have a number75RR wrote: ↑Thu Nov 19, 2020 2:12 pmPaul James is my go to person on COL as wellLouiseScot wrote: ↑Thu Nov 19, 2020 12:32 pmActually, in the end for the COL I just used an iris diaphragm plus a an opaque central stop (as used for dark field) which I'd previously 3D printed. It might be better with a slightly smaller stop than I used. I'll have to experiment to see what works best. There is some info about COL here:
http://www.microscopy-uk.org.uk/mag/artdec02/pjcol.html
Don't know if this will be usefull to you, they increase in size by 1mm
download/file.php?id=335
Louise
A Nikon CF plan 20x; A Swift 380T; A DIY infinity corrected focus rail system with a 40x/0.65 Olympus Plan, a 10x/0.30 Amscope Plan Fluor, and a 20x/0.75 Nikon Plan Apo