403
Machine Learning (mander.xyz)
all 25 comments
sorted by: hot top controversial new old
[-] megopie 60 points 1 month ago

Hell, doesn’t really work in temperate zones ether to be honest. It gets common house/yard plants well but if you go into actual wild-ish areas it will give you 5 different answers from five different angles of the same plant.

[-] nik9000@programming.dev 39 points 1 month ago

I downloaded google lense a while back to identify a mushroom. It was pretty and I was curious. After installing and taking the picture it replied.... "Mushroom."

The second image said false widow's death wish or something metal as hell.

[-] Alwaysnownevernotme@lemmy.world 19 points 1 month ago

I would literally code it to identify every mushroom as a destroying angel or some such. Not worth the liability.

[-] megopie 7 points 1 month ago* (last edited 1 month ago)

Yah, mushrooms are not that hard to identify if you know what to look for. I worry people who do not know what to look for will be far too confident with such an imperfect tool.

Especially with the growth of demand for foraged mushrooms in restaurants and supermarkets. It’s big money, and I have no doubt some “enterprising” people are going to get people hurt by trusting these tools too much.

[-] Honytawk@lemmy.zip 11 points 1 month ago

And it would be less of a problem if it showed all options that are likely, instead of just the one it thinks is most likely.

“Haha Google says this is safe, I’ll try it”

“Why is my entire skin numb”

[-] SapphironZA@sh.itjust.works 21 points 1 month ago

Same here for the southern hemisphere. Google lens has no clue.

[-] lemming@sh.itjust.works 17 points 1 month ago

How does Plantnet fare in tropics?

[-] unexposedhazard@discuss.tchncs.de 19 points 1 month ago

I mean its just a matter of total available data points. The more images people take and upload, the more material they have to train their models. And obviously there will be way less people running around the tropics taking pictures.

[-] fishpen0@lemmy.world 28 points 1 month ago

It’s honestly way more about plant diversity. There are a million different plants in like a ten square mile area that all look exactly like an aloe and are related. The only way to differentiate them is by hyper obscure differences like their root structure and what their sap consists of.

You don’t even need to be in the proper tropics. Walk around San Diego with a plant id app and watch it spit out a different name for the same palm tree over and over because there are actually hundreds of varietals of palm with similar extremely complex identification processes. Some with toxic fruit and some with edible fruit that look the same.

[-] unexposedhazard@discuss.tchncs.de 11 points 1 month ago

I mean for those plants the model should be trained to spit out the next highest common denominator / family instead of the specific species. I would love to get a reply like "this could be any of the following species" instead of "im 23.232% sure that its this species"

[-] stiephelando@discuss.tchncs.de 6 points 1 month ago

Seek does that. That's the reason it's my go to ID app

[-] Swedneck@discuss.tchncs.de 2 points 1 month ago

the full inaturalist app also has you upload the observation and likely get suggested IDs from other people

[-] emergencyfood@sh.itjust.works 3 points 1 month ago

I mean for those plants the model should be trained to spit out the next highest common denominator / family instead of the specific species.

Most people are going to take photos of the leaves, stem or at best the outside of the flowers. These are rarely conserved within families. You'll need the arrangement of the four floral whorls to name a family and expect any degree of accuracy. And that's assuming your plant is an angiosperm.

[-] BastingChemina@slrpnk.net 3 points 1 month ago

From my experience it's quite good in the Caribbean and it's getting better.

[-] lemming@sh.itjust.works 1 points 1 month ago

Cool, thanks for the info!

[-] Swallowtail@beehaw.org 9 points 1 month ago

Try iNaturalist, it works pretty well. Also, learn plant morphology, makes it easier to narrow things down when you get a couple suggestions within the same genus or family.

[-] Leate_Wonceslace@lemmy.dbzer0.com 5 points 1 month ago

As someone who uses iNaturalist and has a plant friend, I can confirm.

[-] Swedneck@discuss.tchncs.de 2 points 1 month ago

using inaturalist and uploading your observations also contributes to science :)

[-] flora_explora@beehaw.org 2 points 1 month ago* (last edited 1 month ago)

Indeed, basic plant morphology knowledge plus some local Floras and iNaturalist worked out quite well for me in the tropics. There are also so many people that know plants on iNat. You only get into trouble if you try to ID rare species, but that's also the case in the temperate zones.

[-] iAvicenna@lemmy.world 5 points 1 month ago

pictures you can hear

[-] Rubanski@lemm.ee 3 points 1 month ago

"picture this" is a terrible name choice, but a really good app for identifying plants

[-] moosetwin@lemmy.dbzer0.com 3 points 1 month ago

haha this is going to kill someone

[-] Spoilt@jlai.lu 2 points 1 month ago

My brain read "pants", I didn't understand the connection with latitude... Thanks, brain.

this post was submitted on 23 Jul 2024
403 points (100.0% liked)

Science Memes

10207 readers
2834 users here now

Welcome to c/science_memes @ Mander.xyz!

A place for majestic STEMLORD peacocking, as well as memes about the realities of working in a lab.



Rules

  1. Don't throw mud. Behave like an intellectual and remember the human.
  2. Keep it rooted (on topic).
  3. No spam.
  4. Infographics welcome, get schooled.


Research Committee

Other Mander Communities

Science and Research

Biology and Life Sciences

Physical Sciences

Humanities and Social Sciences

Practical and Applied Sciences

Memes

Miscellaneous

founded 2 years ago
MODERATORS