Flag or otherwise mark taxon photos whose CID doesn't match taxon

[deboas]

Three more examples of this problem from this thread - further evidence that this feature request would be helpful:

@hkibak

What does one do when the lead photo on a species page is wrong, in the “Needs ID” category, and not subject to “Community Taxon?” It is a common species in Mexico and the Southwest US with over 4,600 observations, and there is evidence folks are misidentifying their observations because of this photo.

https://www.inaturalist.org/taxa/68416-Cylindropuntia-ramosissima

@fubberpish

I had noticed the lead photo for Naso annulatus was actually Naso unicornis, but wasn’t sure what to do about it. thanks to this post, ive been able to change the lead photo to the right species :)

@annkatrinrose

I only fairly recently started replacing obviously wrong photos on iNat taxon pages. I think it’s not immediately clear that this is an option available to everyone (which may be a good thing). The one that made me look into it was a common Eastern NA woodland plant with over 60,000 observations on iNat that was illustrated with an obviously wrong image of some cultivated plant in a garden in England taken from Flickr, not even iNat.

[jeanphilippeb]

@deboas

source_of_taxon_photo=true

That’s fine, as a filter, because extremely few observations would have this status, in comparison with the whole set of observations. As a consequence, no need for a background process.

[deboas]

Another example of a taxon photo that does not match the ID of the source observation. Any update on the suggestions made to address this issue, @tiwane?

[matthewvosper]

I also think this is a good idea. It would need to be run once on every observation that provides a taxon photo, but thereafter it would only need to be triggered when a taxon photo is changed, or when the CID of a taxon-photo-providing observation changes (not every time an ID is added). A flag seems the best approach because it would be searchable.

[annkatrinrose]

A lot of the taxon photos I’ve run into recently that weren’t actually showing the taxon they were supposed to illustrate turned out to be older pictures that were pulled in from Flickr at some point. In those instances, there is no CID since the images aren’t even on iNaturalist. I’m not sure how that would fit into this check - flag all non-iNat pictures to be verified? Many are good and excellent photographs, but I’ve also found quite a few garden cultivars, hybrids, or just plain wrong ones that came from Flickr.

[rupertclayton]

I’m seeing that, too. This was my suggestion for how to handle Flickr images (from earlier in the thread):

Given that the need for off-domain image sources is now much reduced, I think it would also be appropriate to have a similar yellow warning for non-iNat sourced images:

This image was chosen from [Flickr/EOL/Wikimedia Commons] to illustrate the species Eleustrine latifolia. We can’t be sure whether the image creator applied the correct identification and you may want to consider replacing this image with one from a Research Grade iNaturalist observation of this species.

[DianaStuder]

I wonder how many taxon photos are non iNat (Flickr) based.
If we could access a list, we could work thru a chosen section.

[matthewvosper]

Usually these are applied to taxa with few iNat observations I think. Perhaps only flag these once the taxon reaches a certian number of RG observations?

[tiwane]

No, not yet.

[DianaStuder]

There are also lots of taxa with one default, not very helpful, image. Or the chosen default is Walking with My Dog, or with My Hiking Group. No, thank you.
My autopilot works thru the taxa I meet. When I need to.

[rupertclayton]

I agree that Flickr taxon images are a lot more common for less-observed taxa. That’s not surprising as more-observed taxa have more people paying attention to them and more iNat images to choose from. But I don’t think we really need to distinguish these within the logic I’m suggesting to add text like this:

This image was chosen from [Flickr/EOL/Wikimedia Commons] to illustrate the species Eleustrine latifolia . We can’t be sure whether the image creator applied the correct identification and you may want to consider replacing this image with one from a Research Grade iNaturalist observation of this species.

If that warning appears on a photo of a fairly common species, someone should be able to find plenty of iNat photos with validated IDs as replacements. On a rarer species, there may be one or two iNat photos (still worth substituting) or there may be none at all, in which case it’s still worth alerting users that the taxon photo is unvalidated.

[jasonhernandez74]

@rupertclayton

That’s not surprising as more-observed taxa have more people paying attention to them and more iNat images to choose from.

And more frequent changes. Every time I start doing IDs, I wonder, “What’s the default image for ‘Beetles’ today?” There are probably as many opinions as to which beetle picture best embodies ‘Beetles’ as there are iNat users.

[wildwestnature]

we have noticed that when we add a photo to the taxon curation (on PC using Chrome), even when we try to put that new photo at the end of the list of 12 maximum photos, it almost always goes to the front and replaces the default photo. We then drag it from the default (which restores the prior default photo). Maybe it’s user error. We wonder if this could contribute to inadvertent changes in default photo?

(When we are contributing to curation, it has never been to alter the default photo but instead is to add other life stages or sexes (ie bird chicks and female of sp). (Note: And we are not adding our own photos when curating))

[jeanphilippeb]

I use this project for restoring the photos in a taxon page:
https://www.inaturalist.org/projects/taxon-page

I you wish to use it for other taxa, become a member and ask to become a curator.

[deboas]

Two more Flickr photos that have been used as taxon photos for the wrong species: https://forum.inaturalist.org/t/photos-from-flickr-that-are-incorrectly-identified/38812

[deboas]

Another example of taxon photos of species that don’t match the taxon page, highlighting the need for this feature request: https://forum.inaturalist.org/t/how-do-we-change-a-wrong-photo-showing-the-whole-species/39224

[deboas]

And another. Erroneously identified photo from Flickr appearing on species page: https://www.inaturalist.org/flags/610339

[tiwane]

Discussed this request earlier in the week. We don’t want to add more design or generate flags for these photos - we’d want to just remove them from the taxa they’re associated with. We figure the script would run once a month or so, for starters. Photo removal would show up in the taxon’s history page.

To get an idea of how many photos would be removed, we ran two scripts on our test server yesterday.

• one only removed photos which have no observations with a taxon on the same branch as the photo taxon and the taxa are not related by taxon changes (to address some of the concerns brought up here, like with subspecies). This script would remove 157 of the 25823 taxon photos created in the last month (0.6%). Ken-ichi did some spot checking and found that the photos to be removed were good candidates for removal, or were chosen for taxa that need to be synonymized.

• the other didn’t have that branch exception, so it would remove 788 of 25559 taxon photos created in the last month (3%). So unlike the script above, it would remove photos associated with an observation whose community ID is an ancestor of the photo taxon. For example, a photo associated with a Trombidiform mite species but now has a Community ID of Order Trombidiformes would be removed from the species.

(The difference in total taxon photos made in the last month is due to our running the script several hours apart and the “last month” being a relative time.)

Thoughts? I think it’s worth trying, especially with the more conservative script, to see how things look.

And to be clear, this won’t affect photos from sources outside of iNaturalist, of course, like Flickr or Wikimedia.

Fun fact: there are around 1.5 million taxon photos on iNat.

[egordon88]

@tiwane

I think it’s worth trying, especially with the more conservative script, to see how things look.

Seconded

[jwidness]

You would run this on all taxon photos, right? (The “last month” constraint was just for testing?)

I’m in favor, and would prefer tossing more rather than fewer.