Multimodale Koregistrierung an einem Standardmikroskop

[Ramon]

Hello Ramon,
I'm very curious about your device, too.

Hi Holger!

I was waiting for you to contact me. Remember that we agreed upon having a video call so that I can show you the nuances of the gadget.

Look forward hearing from you soon.

Ramon

[Lupus]

Hello Ramon,

the problem is that it is not entirely sufficient for the cut-off filter to cover the entire emission wavelength range of the excitation LED. Its intensity must also be significantly reduced, particularly when the fluorescence yield is low. This is because cameras typically have a significant overlap in the spectral sensitivity of the RGB channels, meaning that blue light can still excite the G and to a lesser extent the R channel. 

In the image below, the top left shows the original photo with UV-violet excitation, whilst on the right the blue channel has been strongly amplified. However, you can see that it has also been exposed. At the bottom left, the red channel has also been amplified. It is also worth noting, however, that the camera's JPG conversion can also artificially create small amounts of colour in an unexposed colour channel if neighbouring pixels of a different colour are more heavily exposed.

And at the bottom right is the RG colour separation, but with the green channel attenuated and the red channel amplified. The red areas could be typical chlorophyll fluorescence. The light green areas are more likely to be scattered light. The red and dark green areas can be seen in the same colour in your brightfield image.



Hubert

[Lupus]

Hello Ramon,

I addressed a similar topic in this article about condenser-free illumination:
https://www.mikroskopie-forum.de/index.php?topic=54614.0 
Using a blue excitation LED in simple dark-field illumination from below and a much more rudimentary red film filter, I attempted to demonstrate chlorophyll fluorescence on a fresh, unstained plant section. It is important to use fresh sections rather than ready-made slides, as chlorophyll degrades.
Here is a section (parsley) under white transmitted light, followed by two fluorescence images. Green should be scattered light.







Hubert

[Ramon]

Hi Hubert,

Yesterday I posted regarding Photo Bleaching as a method for testing for fluorescence in the Jaasmine sample. I had to put it down because there were some errors. I will correct it and put it back in this afternoon/evening.

I addressed a similar topic in this article about condenser-free illumination:
https://www.microscope-forum.de/index.php?topic=54614.0 
 Using a blue excitation LED in simple dark-field illumination from below and a much more rudimentary red film filter, I attempted to demonstrate chlorophyll fluorescence on a fresh, unstained plant section. It is important to use fresh sections rather than ready-made slides, as chlorophyll degrades.
Here is a section (parsley) under white transmitted light, followed by two fluorescence images. Green should be scattered light.

Those are very good images for such a simple setup. And quite humbling because for months now I have been trying to get similar results with the FrTIR gadget. Without success sadly. I imaged mainly moss and algae and the results have been no where near your results. I am going to try again, this time with a fresh stem similar than the one you used.

Ramon

[Lupus]

Hello Ramon,

Chlorophyll is not only bleached by light; there is also biochemical degradation following cutting.

A fundamental problem lies in the spectral sensitivity distribution of a typical camera sensor. The image shows my situation with blue LED excitation. The dotted line illustrates the typical sensitivity curve of an RGB sensor (which may vary significantly in individual cases). My simple red cut-off filter is shown schematically in dark red; the residual transmission is not negligible in the case of blue light. On the far right is the red fluorescence from chlorophyll.

You can see
1. that unblocked residual blue light can also excite the green and less the red pixels, and
2. that the chlorophyll fluorescence lies very far into the long-wavelength red region, where the camera sensor may already be relatively insensitive.

This means that the contrast ratio between the excitation light and the fluorescence light may become very unfavourable. This makes it all the more important to use a fresh plant sample with maximum fluorescence yield, and a good cut-off filter. The Thorlabs 500 nm filter, for example, has a transmission of 0.05 per cent in the blue spectral range – which is not bad but not zero either.
And, of course, minimising stray light from the dark-field illumination is also part of the concept to suppress the excitation light as much as possible.



Hubert

[Ramon]

Hi Hubert

In fairness I should add: the reason I can keep up with your points and reply quickly is that I work through them with an AI assistant. The experiments and the judgement are mine, but it helps me analyse the data and put my thoughts together at a pace I could not manage alone. 

thank you — the spectral diagram is really useful, and both your points are well taken. I want to be honest about my motive here: I have no stake in the "fluorescence" answer. I'd rather end up with a correct "no" than a comfortable "yes."

On biochemical degradation: you are right however this section is a commercial fixed/mounted slide rather than fresh tissue (so little active biochemistry), and the signal that fades most is the green cell-wall autofluorescence (lignin/cutin), which is structural and photobleaches rather than degrades.

On the camera/filter spectral point: Your diagram is blue excitation (~460 nm) with a >600 nm cut-off aimed at chlorophyll red (~680 nm) — which, as you show, lands where the sensor is already insensitive. My run used violet excitation (~400 nm) with a 500 nm long-pass, and the signal I am actually relying on is the green cell-wall autofluorescence (~510–560 nm). That green sits at the sensor's peak sensitivity and just above the filter cut-off, and it is far from a violet excitation the Thorlabs filter blocks hard (~0.05% in the blue, less at 400 nm). So your "far-red, sensor-insensitive" concern really applies to the red/chlorophyll channel — which is exactly the channel I am now treating as unconfirmed.


Thanks again — this genuinely sharpened the work.

Now let me present you the latest version of the Photo Bleaching test:

Photobleaching test — winter jasmine

Frustrated-TIR grazing illumination · violet-UV (~395–405 nm) bleaching · 500 nm long-pass · imaged with the device's own (stable) UVA LEDs. REV3, 24 August 2026.

Purpose

To test whether the emission from the winter-jasmine section is genuine fluorescence or scattered/reflected excitation. Photobleaching distinguishes the two: under a constant excitation, fluorescence fades as the fluorophores are destroyed, whereas scattered or reflected light does not. The same field was bleached in steps and re-imaged, and the region intensities were compared. Excitation is violet-UV (~395–405 nm, UVA SMD5050 LEDs); the emission filter is a 500 nm long-pass.

Set-up and method

This test uses two SEPARATE light sources — this distinction is essential to reading the result:



Image series

The same field imaged at increasing cumulative violet BLEACHING exposure (each panel labelled with total exposure time; each captured under the stable imaging LEDs):



Quantitative analysis

Mean pixel intensity (0–255) in four fixed regions across all six frames: the diffuse background halo, the bright green parenchyma cells, the reddish vascular bundle, and a dark background area far from the specimen.




Plotted against cumulative bleaching dose, the fade is approximately linear (dashed fits):



Interpretation

The imaging excitation was constant. Because each photo was taken with the FrTIR device's own imaging LEDs — ON for only a few seconds, with the bleaching LED off — the light exciting the specimen at every capture was the same.
Under a constant excitation, scattered or reflected light stays constant (it cannot bleach); only fluorescence fades. The signal fell by roughly 25–34%, so the fading part of the signal is genuine fluorescence, not scattered light. The fade is approximately linear with cumulative dose (slope ≈ −0.64 units/min), consistent with photobleaching. Every region fades — including the diffuse halo — so the halo cannot be purely scattered light either.

I wonder if the answer is that we are having fluorescence plus scatter.

I hope to keep up this lively conversation.

Ramon

[Lupus]

Hello Ramon,

I don't actually identify green autofluorescence.
1. The bleaching experiment assumes that areas outside the fluorescence emission spectrum have the same intensity when exposed to the same light, or at least that the measurements are normalised accordingly to this 'background brightness'. However, the green halo also changes in intensity, in a similar ratio to the cell area.
2. Autofluorescence, e.g. from lignin, is brighter in the region of the cell walls or corresponding structures; here, however, the opposite is true, in that the cell interior glows more brightly.

This paper demonstrates lignin fluorescence very clearly (Fig. 5)  https://pmc.ncbi.nlm.nih.gov/articles/PMC7288016/?utm_source=chatgpt.com

Hubert

[Ramon]

Hi Hubert

I don't actually identify green autofluorescence.
1. The bleaching experiment assumes that areas outside the fluorescence emission spectrum have the same intensity when exposed to the same light, or at least that the measurements are normalised accordingly to this 'background brightness'. However, the green halo also changes in intensity, in a similar ratio to the cell area.
2. Autofluorescence, e.g. from lignin, is brighter in the region of the cell walls or corresponding structures; here, however, the opposite is true, in that the cell interior glows more brightly.

This paper demonstrates lignin fluorescence very clearly (Fig. 5)  https://pmc.ncbi.nlm.nih.gov/articles/PMC7288016/?utm_source=chatgpt.com

Give me a day or two to go through this.

I addressed a similar topic in this article about condenser-free illumination:
https://www.mikroskopie-forum.de/index.php?topic=54614.0 
Using a blue excitation LED in simple dark-field illumination from below and a much more rudimentary red film filter, I attempted to demonstrate chlorophyll fluorescence on a fresh, unstained plant section. It is important to use fresh sections rather than ready-made slides, as chlorophyll degrades.
Here is a section (parsley) under white transmitted light, followed by two fluorescence images. Green should be scattered light.

I dont know what I am doing wrong.

Sample: section of stem of a fresh plant

Excitation: cheap blue LED with a green tail (leaks green)

Barrier filter: Throlabs 500nm LPF

Illuminated through the FrTIR device (true grazing angles because we do away with the air/glass interface)




No sign of red emission from chlorophyll

Ramon

[Lupus]

Hello Ramon,

Excitation: cheap blue LED with a green tail (leaks green)

What exactly does that mean? Do you have the emission spectrum from the data sheet?

If the spectral distribution is similar to that in this graph (the yellow line corresponds to your block filter), then B, G and R will, of course, be relatively strongly illuminated by the excitation light, compared with the chlorophyll fluorescence at the less sensitive end of the of the R channel of the camera sensor.



How long was the exposure time for this photograph?

You can see from the histogram of your image (left) that the blue component is also very high. On the right, I've created a green-red colour separation and brightened the red component. So you can already see the chlorophyll fluorescence.
Is the plant section not a cross-section, but rather a stem cut from the side?



Hubert

[mikropit]

Es wäre schon schöner, wenn in einem deutschen Forum auf deutsch kommuniziert würde.
 Ich bekomme zwar auf englisch mein Essen, aber mit diesen Texten sind sicher manche überfordert. Jedes mal mit einem Übersetzungsprogramm übersetzen, ist halt auch nicht ganz zielführend.
Peter Mikropit

[Lupus]

Hallo Peter,

das sind eigentlich Ausnahmefälle, hier ist es ein spanischer Forumsteilnehmer.
Es ist doch ganz einfach: Rechte Maustaste - "Übersetzen in Deutsch" anklicken. Dann erscheint der ganze Thread auf Deutsch.

Hubert

[mikropit]

Hallo, Hubert, diese Möglichkeit kannte ich nicht. Vielen Dank. Ja das ist zumutbar. Nur die Diagramme werden nicht übersetzt.
Peter Mikropit

[Ramon]

Es wäre schon schöner, wenn in einem deutschen Forum auf deutsch kommuniziert würde.
 Ich bekomme zwar auf englisch mein Essen, aber mit diesen Texten sind sicher manche überfordert. Jedes mal mit einem Übersetzungsprogramm übersetzen, ist halt auch nicht ganz zielführend.
Peter Mikropit

Hallo Peter,

du hast völlig recht — aber leider spreche ich kein Deutsch. Kurioserweise ist mein bester Freund seit 30 Jahren Deutscher, und ich habe viele Weihnachtsferien in der schönen Stadt Wolfenbüttel verbracht — ich habe also eine echte Zuneigung zur Sprache und zum Land, auch wenn davon nie wirklich etwas auf meinen Wortschatz abgefärbt hat!

Der eingebaute Übersetzer des Forums ist allerdings wirklich gut. Ich verlasse mich selbst darauf, um jede deutsche Antwort in diesem Thread zu lesen, und ich finde ihn zuverlässig genug, um der Diskussion recht genau zu folgen — einschließlich Huberts detaillierter technischer Punkte, die nicht einfach sind.

Ganz allgemein: Ich bin seit fünf Monaten im Mikroskopie-Forum und auf Photomacrography.net aktiv, und ohne das Feedback aus diesen Gemeinschaften — unter anderem von Hubert, Spektrum und Jürgen — wäre dieses Projekt noch in der Steinzeit. Ich bin zutiefst dankbar und hoffe, dass ich hier weiter dazulernen darf.

Wenn es dir lieber ist, übersetze ich gerne alle meine Beiträge ins Deutsche, bevor ich sie poste. Sag mir einfach Bescheid.

Viele Grüße,
Ramon

das sind eigentlich Ausnahmefälle, hier ist es ein spanischer Forumsteilnehmer.
Es ist doch ganz einfach: Rechte Maustaste - "Übersetzen in Deutsch" anklicken. Dann erscheint der ganze Thread auf Deutsch.

Vielen Dank, Hubert, dass du ein gutes Wort für mich eingelegt hast. Ich gehe gerade unseren Austausch zur Winterjasmin-Probe noch einmal durch. Ich muss dir sagen: Die KI stimmt deinem letzten Beitrag zum Jasmin inzwischen zu 100 % zu. Gib mir bitte etwas Zeit, dann antworte ich ausführlich.

Ramon

[Ramon]

What exactly does that mean? Do you have the emission spectrum from the data sheet?

Hi Hubert,

It means that my blue LEDs emmit unwanted green light.

I dont have datasheet for the blue LEDs. I bought them on Amazon and not much info was available. All the info I have on them is what the AI was able to dig out by the reference. That info being: Wavelength   460 nm (nominal, blue), LED package   SMD 2835 (single-die, monochromatic blue — no phosphor).

You're right about the geometry — it's a stem cut from the side (longitudinal), not a transverse cross-section.

How long was the exposure time for this photograph?

The exposure was long — the software's Exposure setting was ~4550, near its maximum (~5100).

Your spectral point is well taken: with the 500 nm long-pass, the passband opens right where the green channel is most sensitive (~530 nm), so anything in the 500–560 band dominates, while the chlorophyll fluorescence at ~680 nm is faint and sits where the sensor is already going insensitive. The red is there, but swamped.

I always knew that my blue LEDs had a green tail (they emmit green). Putting together the green leak with the passband openning right where the green channel is most sensitive plus a chlorophyll fluorescence at ~680 nm where the sensor is already going insensitive...Well, its easy to understand why I dont see red.

Regarding the "green stuff" in the Jasmine. Please bear with me, I am still working on it. I just discovered that my Violet-UV LEDs are also leaking green light. I am building a DIY spectrometer to confirm it.

Just one question regarding the Jasmine. On the green halo, how could it be scattered light if there are no cell structures there that can scatter?

Ramon

[Lupus]

Hello Ramon,

It means that my blue LEDs emmit unwanted green light.

It is clear that a blue LED also emits longer-wavelength light tending towards the green. My graph shows the emission spectrum from the datasheet of my Nichia LED, which has a peak wavelength of 463 nm; this applies similarly to any LED. As the fluorescence efficiency is often only 1 per cent, or even less, unless special fluorescent dyes are used, any external light of this magnitude masks the fluorescence.

The UV LED also has a similar emission spectrum of comparable relative width, but I think this is barely relevant in the long-wavelength range >500 nm when the emission peak lies 100 nm away.

My question about the exposure time refers to the fact that a very long exposure time could be affected by stray light from the room lighting. Your lighting setup also picks up ambient light; if the room isn't darkened, this can also become a factor. I would carry out a blank test without LED lighting using the same exposure time.

Just one question regarding the Jasmine. On the green halo, how could it be scattered light if there are no cell structures there that can scatter?

It's clear that this is scattered light just because there are no fluorescent plant structures there. Perhaps there are scattering particles in the embedding medium of the specimen. These could be illuminated by emissions scattered horizontally from the nearby specimen. I consider this commercial specimen unsuitable for conducting fluorescence experiments. Your own plant section also shows – as far as can be seen in the image – no scattered light outside the section.

Regarding your fresh plant section: One disadvantage of this longitudinal, surface-cut section of the plant, compared with a cross-section, is that the chlorophyll autofluorescence is overlapped by the green scattered light from the non-chlorophyll plant cells lying behind it.

Hubert