A Microsoft researcher has created software for identifying flower species:
The project came about after random cross-pollination between Microsoft Research Asia chief researcher Yong Rui and botanists at the Chinese Academy of Sciences. Rui told the academics Redmond could use image-matching to help the botanists identify the spread of flowers throughout China, albeit with some pruning of its vast image banks.
Microsoft Research Asia senior research program manager Guobin Wu says a 20-layer deep convolutional neural network was cultivated alongside learnable filters to identify slight variations between flowers. "During the forward pass, each filter is convolved across the width and height of the input volume, computing the dot product between the entries of the filter and the input," Wu says.
Some 800,000 flower snaps were planted into the UC Berkely and open source Caffe deep learning network, leading to an impressive 90 percent species identification accuracy rate.
(Score: 2) by VLM on Tuesday July 26 2016, @05:48PM
I'm pretty sure there's classified classifiers that are really good at detecting certain unlicensed pharmaceutical flowers and that this goes way back to the old days of computation.
In non classified areas I'm pretty sure there was a big DNR project decades ago to track the spread of invasive weeds in inland freshwater lakes.
(Score: 0) by Anonymous Coward on Tuesday July 26 2016, @06:32PM
Well hopefully this plant recognition software will nip the invasive weed problem in the bud. Wouldn't want to beat around the bush. Or perhaps Microsoft is just leading us up the garden path.
(Score: 4, Interesting) by DECbot on Tuesday July 26 2016, @06:02PM
After I RTFS, it was a little disappointing to learn that MS Research + Chinese Academy of Sciences didn't genetically engineer an intelligent flower and an automated Turning tests for flowers. Automated flower facial recognition just isn't as nearly excited after dreaming about smart flowers. My desire for smart flowers was so strong, I had to read the summary twice to fully grok what was being said, and boy, was that second read a real letdown.
cats~$ sudo chown -R us /home/base
(Score: 0) by Anonymous Coward on Tuesday July 26 2016, @10:37PM
I had to read the summary twice to fully grok what was being said, and boy, was that second read a real letdown.
Well the very first word in this post was Microsoft, what did you expect...
(Score: 0) by Anonymous Coward on Tuesday July 26 2016, @06:06PM
Why does the system ignore dumb flowers?
(Score: 2) by bob_super on Tuesday July 26 2016, @06:15PM
it's more complex than the title suggests: it needs to target flowers which are smart enough to have disposable income but dumb enough to react positively to ads...
(Score: 0) by Anonymous Coward on Tuesday July 26 2016, @06:28PM
Once they do that I suppose it's all guns and roses from there
(Score: 2) by FatPhil on Tuesday July 26 2016, @08:31PM
Great minds discuss ideas; average minds discuss events; small minds discuss people; the smallest discuss themselves
(Score: 2) by bradley13 on Tuesday July 26 2016, @06:53PM
The article implies that this system is (will be?) available to hobbiests, but I can't find a link or any further information. Anyone have better luck?
Everyone is somebody else's weirdo.
(Score: 3, Informative) by mcgrew on Tuesday July 26 2016, @07:10PM
Ubergizmo says [ubergizmo.com] "The app doesn’t actually exist yet, but if and when they do release this engine, perhaps botany hobbyists might be interested in checking it out." I doubt MS will have versions for Android and Apple, wanting to sell their own OS, and if it is it won't be free. And also you won't be able to use the database to build your own because every one of those photos are copyrighted.
mcgrewbooks.com mcgrew.info nooze.org
(Score: 2) by RamiK on Tuesday July 26 2016, @07:23PM
Tags all the edible and hallucinogenics shrooms and flowers.
Game mode: 1/6 tags incorrectly labels hallucinogenics as edibles.
Survival mode: You in the forest for weeks on end living off the land.
Russian roulette mode: 1/6 tags labels poisonous as edibles.
Microsoft Labs: Innovating.
compiling...
(Score: 0) by Anonymous Coward on Tuesday July 26 2016, @10:56PM
... is a system to classify snowflakes..err.. millenials
oh. We have one? They post it themselves? Wow. Okay. Including their sex liv..oh.. pornhub.. right
moving on
(Score: 2) by Knowledge Troll on Wednesday July 27 2016, @02:24PM