search

nvs-vocabs / S06

BODC parameter semantic model parameter entity names
1 stars 1 forks source link

NTR: Request for Area (in digital image) (BODCNVS-1876) #21

Closed PatriciaCabrera closed 1 year ago

PatriciaCabrera commented 3 years ago

Required

Term name (PrefLabel)

Area (in digital image)

Definition

The total surface area (length*length) of an object captured in a digital image and normally expressed in µm2 or mm2.

Sources/references

http://thesauri.lifewatchitaly.eu/PhytoTraits/index.php?tema=16&/surface-area Gorsky et. al. 2010, https://doi.org/10.1093/plankt/fbp124

Optional

Mapping to external terminologies

http://vocab.nerc.ac.uk/collection/P24/current/AREA/1/

ORCID

https://orcid.org/0000-0003-3293-5612

gwemon commented 3 years ago

@PatriciaCabrera I am unsure about this one. If it is expressed in um2 or mm2 then I guess this is an estimate of the real size of the object? If this is the case then we should not specify (in digital image) in the property. However I can understand that simply stating "Area of biological entity specified elsewhere" is ambiguous and not very accurate. Are you aware of the terminology used in digital image processing for this kind of, what I imagine to be, a derived parameter? is it simply the area measured as seen in the digital image in pixel^2 and then extrapolated to real size in um^2 or mm^2?

PatriciaCabrera commented 3 years ago

@gwemon adding here Jean-Olivier Irisson @jiho who can help in answering this question.

jiho commented 3 years ago

@gwemon Indeed, the measurement is in pixel^2 and then converted. We just discussed this and realised that, for consistency with Length (http://vocab.nerc.ac.uk/collection/P01/current/LGPIXEL1/) and Width, it would be better to keep it in pixels here too.

Then we have asked experts in digital imaging for the most canonical terminology regarding what we commonly call "area". We will get back to you with an updated specification for the term ASAP.

PatriciaCabrera commented 3 years ago

@jiho & @gwemon Now that Length (expressed as pixels) has been created (http://vocab.nerc.ac.uk/collection/P01/current/LGPIXEL1/), I wonder if it is more clear to define Area and to semantically link these terms?

jiho commented 2 years ago

It took me a long time to get back on this but I did not forget (well I did but my TODO software reminded me!).

So it looks like a common and well understood + precise term would be "silhouette area" (of an object in digital image).

PatriciaCabrera commented 1 year ago

@jiho do we still need this: "silhouette area" (of an object in digital image)?

jiho commented 1 year ago

I think it would be useful, yes. We have length and width in pixels, we can have area in pixels too. These are some of the most ecologically important measurements we do on images. From OBIS' point of view, they will only be useful for datasets provided at individual level but some people are starting to provide those so it is relevant.

PatriciaCabrera commented 1 year ago

Oh but I see we have Area in pixels: https://vocab.nerc.ac.uk/collection/P01/current/ARPIXEL2/.

gwemon commented 1 year ago

@PatriciaCabrera This code (https://vocab.nerc.ac.uk/collection/P01/current/ARPIXEL2/) is the area of the pixel itself rather than the Area of an object in pixel^2. We can create a new S06 for "Area (expressed as pixels)" and combine it to create a new P01 with Preflabel: "Area (expressed as pixels) of biological entity specified elsewhere".

PatriciaCabrera commented 1 year ago

@gwemon True! I read it too quickly. Indeed we need: "Area (expressed as pixels) of biological entity specified elsewhere". Can you please create it? Thanks!

jakjonbodc commented 1 year ago

Created new S06 term for 'Area (expressed as pixels)' - S0600316

http://vocab.nerc.ac.uk/collection/S06/current/S0600316/

jakjonbodc commented 1 year ago

Created new P01 term for 'Area (expressed as pixels) of biological entity specified elsewhere' - APIXBI01

http://vocab.nerc.ac.uk/collection/P01/current/APIXBI01/