Artificial intelligence identifies plant species by looking at them

Originally published at:

I’m not actually behind the idea of replacing humans in this endeavor (I think society should adult up and fund the workings/upkeep of the world that’s now become its zoo), but I do wonder if this were launched into the field (where problems like cost and dangerousness are issues), a human shepherd or two could guide the work of a small flock of AI bots in doing this kind of ID. Plants, bugs, fungus, and other tiny things could be so localized you might never know about a species otherwise. Lol, the team would also need an IT person, definitely.


Can someone run this on the Voynich manuscript? Please?


Yet another big step in Jeff Sessions’ war on weed. Can the Terminator be far behind?

Just looking at them?


Me too.


Any word on whether it reserves a special intensity of gaze for bananas?


(Having read only the headline,) I think that is mostly how humans do it too.

1 Like

The new Turing test!

It’s just that this is what we used to call an expert system, not an AI. Because this isn’t AI.

Did you have a look at the paper? They used a deep learning algorithm. Despite not being intelligent in any human sense, humans call these algorithms AI. ^^

Oh, and BTW:

All datasets were normalized to an uniform size of 256 by 256 pixels without any other type of pre-processing. This is the current state-of-the-art resolution as deep learning models are intensive in computing.

Holy shit, Batman - even with ultrasonic vision I wouldn’t be able to ID at this resolution.

I’ll have to look at the details later, but this is definitely interesting. Also:

[W]e did show that it is not beneficial to do transfer learning from herbarium data to leaf scan pictures and it is even counterproductive to do transfer learning from herbarium data to field images.

Not unexpected, but also interesting. Especially that they even tried on fruits. Leaves, ok. But fruits? :banana: Noone would identify a banana after looking at a flattened, dried specimen of the fruit, would they?

I remember the Microsoft research division training an AI to identify plant pics from China.

I started to think that an AI expert system for identifying edible mushrooms in the wild would be great, but then I realized: There’s no downside for the AI if it guesses wrong.

Deep learning is an advanced expert system on good hardware.
Still no AI.

I want that for my weed plucking robot. Does it run on Raspberry Pi?

1 Like

Holy Chao! Just a few days ago I was pondering if the technology was ready for that, when my girlfriend and I were looking at some bushes we couldn’t identify. My guess was “not there yet, but close.” I also pondered the interest of crowdsourcing a database of plant pictures for the purpose of identifying them in the wild from a smartphone.

This topic was automatically closed after 5 days. New replies are no longer allowed.