Identify Plants with an App • Hacker News Board

Re: Identify Plants with an App

#31

Post by dharma1 » Tue, Apr 25, 2017, 8:32 AM UTC

I did this for common plant diseases using Tensorflow on Android when learning how to use TF last year. Found the accuracy to be not that great with real world images that are not in the dataset (with a dataset of 50,000 images from PlantVillage). I think there are just too many visual similarities between various plants (especially when taken from various angles) that the network doesn't focus on the right things. Did a short writeup here:

http://helminen.co/plant-disease

It could work for a subset of plants, or potentially with a much larger training set - but I think NIR spectral/hyperspectral imaging would be the way forward here with more differentiating data points.

Re: Identify Plants with an App

#32

Post by 2845197541 » Tue, Apr 25, 2017, 8:33 AM UTC

Finally, I can identify mushrooms that will kill me and eat the ones that won't.

Re: Identify Plants with an App

#33

Post by oneeyedpigeon » Tue, Apr 25, 2017, 8:37 AM UTC

dharma1 wrote: post #31

I did this for common plant diseases using Tensorflow on Android when learning how to use TF last year. Found the accuracy to be not that great with real world images that are not in the dataset (with a dataset of 50,000 images from PlantVillage). I think there are just too many visual similarities between various plants (especially when taken from various angles) that the network doesn't focus on the right things. D…

Did you use inputs other than just image for identification? I would have thought that adding location and time of year alone could significantly improve results.

Re: Identify Plants with an App

#34

Post by oneeyedpigeon » Tue, Apr 25, 2017, 8:38 AM UTC

2845197541 wrote: post #32

Finally, I can identify mushrooms that will kill me and eat the ones that won't.

I hope, for the author's sake, the app has robust Ts&Cs ...

Re: Identify Plants with an App

#35

Post by ovi256 » Tue, Apr 25, 2017, 8:40 AM UTC

valdiorn wrote: post #26

> Fast forward to today - the tech is available and the team is assembled to make PlantSnap a reality. With a beta version completed, and 250,000 images in the database 250,000 images doesn't sound like a large enough training set to be effective on anything but the most common plants.

To help others understand why, the "Deep Learning" book, which summarizes current state-of-the-art DL, advises having at least 5000 samples (images) per class for OK performance (equal to non-DL approaches) and 50k to 100k for state-of-the-art. A class here would be a plant species.

So even with 5k samples, the 250k image corpus would only have 50 species, using this rule of thumb. A good engineer could pick up DL and build a system that performs to this standard, because the tricks are all written down in the literature.

If they do better, they either exploit unpublished methods, or researched those methods themselves, with their researchers.

Re: Identify Plants with an App

#36

Post by dharma1 » Tue, Apr 25, 2017, 8:41 AM UTC

oneeyedpigeon wrote: post #33

dharma1 wrote: post #31

I did this for common plant diseases using Tensorflow on Android when learning how to use TF last year. Found the accuracy to be not that great with real world images that are not in the dataset (with a dataset of 50,000 images from PlantVillage). I think there are just too many visual similarities between various plants (especially when taken from various angles) that the network doesn't focus on the right things. D…

Did you use inputs other than just image for identification? I would have thought that adding location and time of year alone could significantly improve results.

just images. Location and time of the year both great ideas, but that data wasn't available in the training set

Re: Identify Plants with an App

#37

Post by rorykoehler » Tue, Apr 25, 2017, 9:00 AM UTC

nr23 wrote: post #27

I use this android app for recognizing plants by picture: http://identify.plantnet-project.org/ It's made by french researchers I think. It doesn't work perfectly but I did identify a lot of plants I don't know with it. You can snap multiple pictures of the same plant, for example, 2 pictures of the leaves, 1 of a flower and 1 of the bark, and then use the combination to search. You can also submit your observations…

It doesn't work very well in Singapore at all. I guess the training set didn't include much from this part of the world.

Re: Identify Plants with an App

#38

Post by madshiva » Tue, Apr 25, 2017, 9:03 AM UTC

aphextron wrote: post #29

madshiva wrote: post #17

I want do the same for mushroom, because they are hard to recognize. Google images works quite fine.

Visual identification of mushrooms can be deadly. Always do a proper spore print and have it analyzed by a mycologist.

wut? I know it can be deadly but it's your responsibility to eat it or not. I think about tell information, like you want for other type of plant too, you want know if it's the right time to cut it, etc. nothing about eating or not. I don't need really an apps for finding some morels.

Re: Identify Plants with an App

#39

Post by pingou » Tue, Apr 25, 2017, 9:21 AM UTC

madshiva wrote: post #17

I want do the same for mushroom, because they are hard to recognize. Google images works quite fine.

I did it, but the results can be pretty disappointing, I'm still working on improving them:

https://play.google.com/store/apps/details?id=com.pingou.cha...

https://itunes.apple.com/us/app/champignouf/id1227854971

ps: any improvement suggestion is welcome!

Re: Identify Plants with an App

#40

Post by yasth » Tue, Apr 25, 2017, 9:52 AM UTC

folli wrote: post #25

Definitely an interesting concept. There's a bit more information on the indiegogo website [1]. Their problems they mention seem to be quite standard for image recognition (scale- and perspective variance), however I could imagine that this could be quite disastrous for plant recognition, given the massive diversity of plants: e.g. a leaf viewed from the side could equally well be a more narrow leaf from a different…

Honestly they don't seem to be that close to cracking it, like at all.

They may be using camera phone images, but they are well framed images of very distinctive features. Trying to recognize a literal tree in a forest is going to be far harder, and recognizing a tree based on a dead leaf (as a photo towards the bottom suggests) can be very very tricky. Also I have to admit to being concerned that adding "any known plant" won't kill accuracy. The more classification endpoints the less likely you are to get a decent result, and already you have to deal with far greater scale differences than are usual (basically it would be like identifying the breed of dog from anything from a full picture to a photo of a single claw).