[NTR] Add resolution definitions

Zack-83 commented 4 months ago

There are four types of data resolution: spatial, spectral, radiometric and temporal. Many instruments can capture one or two types of resolution simultaneously, yet it is rare to find equipment capable of delivering on all four types. This phenomenon is known as the resolution trade-off. Most instruments measure the most commonly used types of resolutions: spatial and spectral. Together, spatial and spectral resolutions allow scientists to quantitatively measure factors such as color, space and detail.

Preferred term labels and textual definitions

Resolution interval (science) == Minimum difference between two measurements so that they can be resolved (i.e., they can be considered different)
- Spatial resolution (microscopy) == Minimum distance between two points so that they can be considered two separate points in a certain microscope with given experimental parameters
- Lateral resolution as angle: $δθ = 1.22\cdotλ/D$
- Lateral resolution as length: $δr = 0.61\cdotλ/NA$ (Abbe formula, derived from the Rayleigh criterion with the Kirchhoff, Debye and paraxial approximation) $= 1.22 λ\cdot(f/\sharp)$
- Axial resolution as length: $δz = 2\cdot n\cdotλ/NA²$
- Spectral resolution (spectroscopy) == Minimum distance between two very narrow peaks so that they can be seen as two separate peak
- Spectral resolution as wavelength $δλ = λ / (n\cdot N) = RF\cdot Δλ\cdot WS / n$
- Spectral resolution as wavenumber $δ\barν$
- Spectral resolution as frequency $δν$
- Spectral resolution as energy $δ E$
- Radiometric resolution == Minimum difference of irradiated energy to which can be (or are) given different "values" (e.g., different levels in a gray scale or in another look-up-table - LUT)
- Temporal resolution == Temporal distance between two frames or measurements (common but somehow improper definition). In techniques where the temporal resolution is not given directly from a temporal distance between subsequent measurements (e.g., pulsed time resolved pump-probe techniques), it can be used more properly as defined above. = 1 / frame rate
Resolution power
- Spectral resolution power $R = λ/δλ = ν/δν$ etc.

Synonyms

...

Textual definition

s. above

Link to CHARISMA Wiki equivalent

https://wiki.charisma.ideaconsult.net/wiki/Item:Q22 Resolution (Raman)
https://wiki.charisma.ideaconsult.net/wiki/Item:Q200 Spectral resolution
https://wiki.charisma.ideaconsult.net/wiki/Item:Q198 Pixel resolution

None of them is exhausting.

Suggested parent term

Please look in the hierarchy in a browser such as the NFDI4Chem TS

Examples

Please provide a more detailed description that examplifies how the term is supposed to be used, in terms of what it is suppose to repesent (e.g. a concept or a relation).

Attribution

If you would like a nanoattribution, please indicate your ORCID id

Sources

https://www.hunterlab.com/blog/what-is-spectral-resolution/ --> https://en.wikipedia.org/wiki/Image_resolution#Types

https://phys.libretexts.org/Bookshelves/Mathematical_Physics_and_Pedagogy/Computational_Physics_(Chong)/11%3A_Discrete_Fourier_Transforms/11.02%3A_Spectral_Resolution_and_Range

https://en.wikipedia.org/wiki/Optical_resolution

http://vikdhillon.staff.shef.ac.uk/teaching/phy217/instruments/phy217_inst_dispersion.html https://www.azom.com/article.aspx?ArticleID=13369 https://www.ossila.com/pages/spectrometer-resolution

StroemPhi commented 3 months ago

For possible reference, there is this resolution term in AFO, which is defined as an iao:information content entity. THe question for us will be if this has to be defined as a measurement datum or a setting datum.

Zack-83 commented 3 months ago

@https://github.com/Stefano-Luin Stiamo cercando di mettere in ordine tra le diverse definizioni di "risoluzione". Vuoi contribuire anche tu?

StroemPhi commented 3 months ago

we will need to look up "sectral dispersion" as defined in CHARISMA, once it is in the https://wiki.charisma.ideaconsult.net/wiki/List_of_all_terms

Stefano-Luin commented 3 months ago

Resolution in science is, in general, the minimum difference between two measurements so that they can be resolved (i.e., they can be considered different). In microscopy, it is the minimum distance between two points so that they can be considered two separate points in a certain microscope with given experimental parameters. In spectroscopy, it is the minimum distance between two very narrow peaks so that they can be seen as two separate peak. In radiometry, it is the minimum difference of irradiated energy to which can be (or are) given different "values" (e.g., different levels in a gray scale or in another look-up-table - LUT). From a temporal point of view, the term resolution is used (somehow improperly), in the case of movies or time lapses in general, to indicate the temporal distance between two frames or measurements. In techniques where the temporal resolution is not given directly from a temporal distance between subsequent measurements (e.g., pulsed time resolved pump-probe techniques), it can be used more properly as defined above. The general exact definition given above does not allow assigning easily an exact value to the resolution, because the resolving power of an instrument in a given measurement would depend on so many things (physical limits, like the diffraction one both spatially and spectrally; signal to noise ratio; a priori knowledge of the sample, etc.). For this reason a "standard" value is often decided, which allows comparing different instruments, and this causes the so-many different definitions that can be found. E.g., in microscopy, a standard value is the distance from the center to the first minimum of the Airy function given the numerical aperture of an objective (NA), which gives the Abbe formula for resolution equal to 0.61/NA. In spectroscopy, when a spectrum is measured, it can be the full width at half maximum of the measured spectrum of strictly monochromatic light (i.e., a spectral line much smaller than the resolution itself). However, there are actually as many definition of “standard resolution” as there are types of instruments (and on “Charisma” I have seen this kind of definition). I found a nice discussion on what is and what is not resolution (at least in microscopy) on the book "Introduction to Confocal Fluorescence Microscopy", Michiel Müller, edited by SPIE press (WA, USA), second edition (2006), page 14 and following. In particular, figure 1.15 illustrates that resolution and sampling should not be confused. According to the Niquist theorem, the more indicated sampling given a certain resolution is two “points” per resolution value.

Stefano-Luin commented 3 months ago

Another thing: the resolution trade-off, for what I read or wrote, is not linked to the fact that you cannot "resolve" more than one or two things with an instrument... It is more linked to the fact that, usually, the better one kind of resolution is, the worse are the others. E.g., in microscopy, to have very high spatial resolution, usually you need to integrate more time or in any case you need more time to acquire an image.

Zack-83 commented 3 months ago

[Uploading Müller - Confocal Fluorescence Microscopy pp 14-23.pdf…]()

Stefano-Luin commented 3 months ago

I'm not sure if the file should be inserted anywhere... Anyway, I signal that pages 14-17 and 22-23 are repeated twice.

I confirm that those are the pages I was referring to in the comment of mine.

Best wishes, Stefano

Il giorno lun 3 giu 2024 alle ore 14:54 Giacomo Lanza < @.***> ha scritto:

Müller - Confocal Fluorescence Microscopy - pp 14-23.pdf https://github.com/user-attachments/files/15534052/Muller.-.Confocal.Fluorescence.Microscopy.-.pp.14-23.pdf

Zack-83 commented 2 months ago

* **Resolution interval (science)** == Minimum difference between two measurements so that they can be resolved (i.e., they can be considered different)

  * **Spatial resolution (microscopy)** == Minimum distance between two points so that they can be considered two separate points in a certain microscope with given experimental parameters

    * Lateral resolution as angle:  δθ=1.22⋅λ/D
    * Lateral resolution as length: δr=0.61⋅λ/NA
      (Abbe formula, derived from the Rayleigh criterion with the Kirchhoff, Debye and paraxial approximation)
    * Axial resolution as length:   ²δz=2⋅n⋅λ/NA²
  * **Spectral resolution (spectroscopy)** == Minimum distance between two very narrow peaks so that they can be seen as two separate peak

    * Spectral resolution as wavelength δλ=λ/(n⋅N)=RF⋅Δλ⋅WS/n
    * Spectral resolution as wavenumber δν¯
    * Spectral resolution as frequency  δν
    * Spectral resolution as energy     δE
  * **Radiometric resolution** == Minimum difference of irradiated energy to which can be (or are) given different "values" (e.g., different levels in a gray scale or in another look-up-table - LUT)
  * **Temporal resolution** == Temporal distance between two frames or measurements (_common but somehow improper definition_). In techniques where the temporal resolution is not given directly from a temporal distance between subsequent measurements (e.g., pulsed time resolved pump-probe techniques), it can be used more properly as defined above.   = 1 / frame rate

Definitions updated. We can talk about whenever you want. Thanks to Stefano!

NFDI4Chem / VibrationalSpectroscopyOntology