[NatureNS] AI and species ID?

DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=hotmail.com;
From: Fritz McEvoy <fritzmcevoy@hotmail.com>
To: Stephen Shaw <srshaw@Dal.Ca>,
Thread-Topic: [NatureNS] AI and species ID?
Thread-Index: AQHUxuVItcdBv2bghkOWa2EHVuDXp6XkQreAgAAOGXmAABUlgIAAUEcAgAAZL8g=
Date: Mon, 18 Feb 2019 04:14:07 +0000
References: <DM6PR18MB250600B7D86A51E6B12F177DB7620@DM6PR18MB2506.namprd18.prod.outlook.com>
Accept-Language: en-CA, en-US
Precedence: bulk
Return-Path: <naturens-mml-owner@chebucto.ns.ca>
Original-Recipient: rfc822;"| (cd /csuite/info/Environment/FNSN/MList; /csuite/lib/arch2html)"

next message in archive
no next message in thread
previous message in archive
previous message in thread
Index of Subjects

arger Or
--_000_DM6PR18MB25062AFA0D05D1823819C023B7630DM6PR18MB2506namp_
Content-Type: text/plain; charset="Windows-1252"
Content-Transfer-Encoding: quoted-printable

Hi Stephen and All,
      My interest is less on birds and more on Lepidoptera identification. =
There are plenty of websites that have excellent Lepidoptera photos and lot=
s of links to species descriptions and, if you have the interest, you can l=
earn the basics of species identification.
      The problem with most sites, like BugGuide and iNaturalist, is that m=
ost of the photos are live and taken with wildly varying conditions and equ=
ipment. In Lepidoptera there are mainly two kinds of photos live and spread=
. I do a variation - non spread specimens - which is more common in Coleopt=
era. I think the real future in AI species identification lies in museum co=
llections and in collections like the Barcode of Life Data System (BOLD), w=
hich are mostly spread specimens where photos can be taken under somewhat c=
ontrolled conditions.
        I agree with you that one of the big problems in identifying insect=
s, particularly micro insects, is that much of the fauna is undescribed or =
was last described 100 years ago under a different name. I recently worked =
with a specimen that is acknowledged by experts as a new species but is lis=
ted under another name because no one has gotten around to describing it an=
d only certain experts on any particular Genus have their descriptions acce=
pted by the Lepidoptera community. I've also encountered some Genera in Nor=
th America where many of their species remain undescribed. As you say it's =
a problem.
        Anyway I don't think AI will be a solution for me in the near futur=
e but I will definitely follow its progression. All the best.
                    Fritz
________________________________
From: naturens-owner@chebucto.ns.ca <naturens-owner@chebucto.ns.ca> on beha=
lf of Stephen Shaw <srshaw@Dal.Ca>
Sent: February 17, 2019 8:36 PM
To: naturens@chebucto.ns.ca
Subject: Re: [NatureNS] AI and species ID?

Re Dave=92s comment, whatever your ultimate interest, the thing you need at=
 some point is a species ID.

The current criterion used in the interesting link Ian provides is for at l=
east 20 good images of a species to be available that have been vetted by e=
xperts, for training the AI network so that it can identify a new image to =
a fairly high level of reliability, presumably from a variety of viewing an=
gles.  Given the popularity of digital cameras this could be useful for N A=
merican birds (most quite large, not too many species, hoards of amateur ob=
servers some of whom are expert photographers).  As for insects, it should =
work also for NS butterflies (few species, all IDs known except perhaps for=
 occasional incoming non-native strays, a few expert photographers, main ID=
 by wing pattern & colour (=3D easy), existing web site).

Re flies mentioned by Fritz, far less likely to happen for such much larger=
 Orders =97 far too many species, many too small to photo without special l=
enses or a microscope, fewer interested observers.  A few fly groups have l=
arger, photogenic species (families like syrphids =97 hoverflies) and BugGu=
ide.net<http://BugGuide.net> has hundreds of images of these (~236 come up =
on the page for the large, common alien hoverfly Eristalis tenax, though ~5=
% of images there are actually other species that have leaked into the E. t=
enax page).   Many flies, like tachinids, a huge group, have species that a=
re identifiable, if at all, by bristle patterns on particular cuticular pla=
tes that would need posed photographs by an expert down a microscope to see=
.

A larger problem is that whole blocks of the larger insect groups are not e=
ven fully or satisfactoriiy classified (or even identified) beyond the fami=
ly or genus level yet, to species.  This is likely to alter only very slowl=
y in the foreseeable future =97 few professional taxonomists, given more se=
emingly attractive career alternatives, just too many unsatisfactory insect=
 groupings which professionals assert need revision.  If you want a fly=92s=
 ID, your best initial bet is BugGuide and ask for help.

Steve (Hfx)
-------------------------------------------------------------

On Feb 17, 2019, at 3:49 PM, Ian Manning <ianmanning4@gmail.com<mailto:ianm=
anning4@gmail.com>> wrote:

Hi Fritz,

In the last six months or so, the inaturalist website added machine learnin=
g functionality to image observations, so now when an image is uploaded it =
will give you suggestions based on their machine learning image recognition=
 model, in the video I sent before around the 3-minute mark, Alex Shepard s=
ays the model can ID to common ancestor (higher level taxon) with an accura=
cy of 93%. Each time an observation is uploaded and vetted as "research gra=
de", that image is added to the image analysis training library (used to tr=
ain the model and make it more accurate down the line).

There's lots of other examples of Computer Science folks using Machine Lear=
ning for this type of thing for citizen science (see Merlin like Rick menti=
oned), and also a bunch of more specific/academic purposes, though as far a=
s I know, none are as accessible or ambitious as iNat. Here's some more inf=
o on the computer vision side of things. https://www.inaturalist.org/pages/=
computer_vision_demo


Cheers,
IM



On Sun, 17 Feb 2019 at 15:07, Fritz McEvoy <fritzmcevoy@hotmail.com<mailto:=
fritzmcevoy@hotmail.com>> wrote:
HI Ian,
     I've been on iNaturalist for a couple of years now and it is a great w=
ebpage. There are lots of sites like iNaturalist online now and they are re=
ally helpful for having others help you identify a species.
     The thing I'm asking about is a program that would allow someone to up=
load a photo and/or maybe some data (size, colour, location and whatever ot=
her ID info you have) of a fly, a leaf or a mouse etc. and have a AI comput=
er program take that photo and/or data and give you its guess as to Genus/s=
pecies identification based on it's algorithms - much like facial recogniti=
on and retina scans work today. All the best.
                        Fritz McEvoy
________________________________
From: naturens-owner@chebucto.ns.ca<mailto:naturens-owner@chebucto.ns.ca> <=
naturens-owner@chebucto.ns.ca<mailto:naturens-owner@chebucto.ns.ca>> on beh=
alf of Ian Manning <ianmanning4@gmail.com<mailto:ianmanning4@gmail.com>>
Sent: February 17, 2019 1:43 PM
To: naturens@chebucto.ns.ca<mailto:naturens@chebucto.ns.ca>
Subject: Re: [NatureNS] AI and species ID?

Try uploading a photo to iNaturalist and seeing what it comes up with. I th=
ink you'll be impressed.

Cheers,
IM


--
We're mapping research in Southwest Nova Scotia.
Link to your research project/publications here<https://goo.gl/forms/1JkoBR=
WIP3Kbvxl53>.



--_000_DM6PR18MB25062AFA0D05D1823819C023B7630DM6PR18MB2506namp_
Content-Type: text/html; charset="Windows-1252"
Content-Transfer-Encoding: quoted-printable

<html>
<head>
<meta http-equiv=3D"Content-Type" content=3D"text/html; charset=3DWindows-1=
252">
<style type=3D"text/css" style=3D"display:none;"> P {margin-top:0;margin-bo=
ttom:0;} </style>
</head>
<body dir=3D"ltr">
<div style=3D"font-family: Calibri, Helvetica, sans-serif; font-size: 12pt;=
 color: rgb(0, 0, 0);">
Hi Stephen and All,</div>
<div style=3D"font-family: Calibri, Helvetica, sans-serif; font-size: 12pt;=
 color: rgb(0, 0, 0);">
&nbsp; &nbsp; &nbsp; My interest is less on birds and more on Lepidoptera i=
dentification. There are plenty of websites that have excellent Lepidoptera=
 photos and lots of links to species descriptions and, if you have the inte=
rest, you can learn the basics of species identification.&nbsp;</div>
<div style=3D"font-family: Calibri, Helvetica, sans-serif; font-size: 12pt;=
 color: rgb(0, 0, 0);">
&nbsp; &nbsp; &nbsp; The problem with most sites, like BugGuide and iNatura=
list, is that most of the photos are live and taken with wildly varying con=
ditions and equipment. In Lepidoptera there are mainly two kinds of photos =
live and spread. I do a variation - non spread
 specimens - which is more common in Coleoptera. I think the real future in=
 AI species identification lies in museum collections and in collections li=
ke the Barcode of Life Data System (BOLD), which are mostly spread specimen=
s where photos can be taken under
 somewhat controlled conditions.</div>
<div style=3D"font-family: Calibri, Helvetica, sans-serif; font-size: 12pt;=
 color: rgb(0, 0, 0);">
&nbsp; &nbsp; &nbsp; &nbsp; I agree with you that one of the big problems i=
n identifying insects, particularly micro insects, is that much of the faun=
a is undescribed or was last described 100 years ago under a different name=
. I recently worked with a specimen that is acknowledged
 by experts as a new species but is listed under another name because no on=
e has gotten around to describing it and only certain experts on any partic=
ular Genus have their descriptions accepted by the Lepidoptera community. I=
've also encountered some Genera
 in North America where many of their species remain undescribed. As you sa=
y it's a problem.</div>
<div style=3D"font-family: Calibri, Helvetica, sans-serif; font-size: 12pt;=
 color: rgb(0, 0, 0);">
&nbsp; &nbsp; &nbsp; &nbsp; Anyway I don't think AI will be a solution for =
me in the near future but I will definitely follow its progression. All the=
 best.</div>
<div style=3D"font-family: Calibri, Helvetica, sans-serif; font-size: 12pt;=
 color: rgb(0, 0, 0);">
&nbsp; &nbsp; &nbsp; &nbsp; &nbsp; &nbsp; &nbsp; &nbsp; &nbsp; &nbsp; Fritz=
</div>
<div id=3D"appendonsend"></div>
<hr style=3D"display:inline-block;width:98%" tabindex=3D"-1">
<div id=3D"divRplyFwdMsg" dir=3D"ltr"><font face=3D"Calibri, sans-serif" st=
yle=3D"font-size:11pt" color=3D"#000000"><b>From:</b> naturens-owner@chebuc=
to.ns.ca &lt;naturens-owner@chebucto.ns.ca&gt; on behalf of Stephen Shaw &l=
t;srshaw@Dal.Ca&gt;<br>
<b>Sent:</b> February 17, 2019 8:36 PM<br>
<b>To:</b> naturens@chebucto.ns.ca<br>
<b>Subject:</b> Re: [NatureNS] AI and species ID?</font>
<div>&nbsp;</div>
</div>
<div style=3D"word-wrap:break-word">Re Dave=92s comment, whatever your ulti=
mate interest, the thing you need at some point is a species ID.
<div>&nbsp;&nbsp;
<div>The current criterion used in the interesting link Ian provides is for=
 at least 20 good images of a species to be available that have been vetted=
 by experts, for training the AI network so that it can identify a new imag=
e to a fairly high level of reliability,
 presumably from a variety of viewing angles. &nbsp;Given the popularity of=
 digital cameras this could be useful for N American birds (most quite larg=
e, not too many species, hoards of amateur observers some of whom are exper=
t photographers). &nbsp;As for insects, it
 should work also for NS butterflies (few species, all IDs known except per=
haps for occasional incoming non-native strays, a few expert photographers,=
 main ID by wing pattern &amp; colour (=3D easy), existing web site). &nbsp=
;
<div><br>
</div>
<div>Re flies mentioned by Fritz, far less likely to happen for such much l=
arger Orders =97 far too many species, many too small to photo without spec=
ial lenses or a microscope, fewer interested observers. &nbsp;A few fly gro=
ups have larger, photogenic species (families
 like syrphids =97 hoverflies) and <a href=3D"http://BugGuide.net">BugGuide=
.net</a> has hundreds of images of these (~236 come up on the page for the =
large, common alien hoverfly Eristalis tenax, though ~5% of images there ar=
e actually other species that have leaked
 into the E. tenax page). &nbsp; Many flies, like tachinids, a huge group, =
have species that are identifiable,&nbsp;if at all, by bristle patterns on =
particular cuticular plates that would need posed photographs by an expert =
down a microscope to see.</div>
<div><br>
</div>
<div>A larger problem is that whole blocks of the larger insect groups are =
not even fully or satisfactoriiy classified (or even identified) beyond the=
 family or genus level yet, to species. &nbsp;This is likely to alter only =
very slowly in the foreseeable future
 =97 few professional taxonomists, given more seemingly attractive career a=
lternatives, just too many unsatisfactory insect groupings which profession=
als assert need revision. &nbsp;If you want a fly=92s ID, your best initial=
 bet is BugGuide and ask for help. &nbsp;</div>
<div><br>
</div>
<div>Steve (Hfx) &nbsp; &nbsp;</div>
<div>-------------------------------------------------------------</div>
<div><br>
<div>
<div>On Feb 17, 2019, at 3:49 PM, Ian Manning &lt;<a href=3D"mailto:ianmann=
ing4@gmail.com">ianmanning4@gmail.com</a>&gt; wrote:</div>
<br class=3D"x_Apple-interchange-newline">
<blockquote type=3D"cite">
<div dir=3D"ltr">
<div dir=3D"ltr">Hi Fritz,
<div><br>
</div>
<div>In the last six months or so, the inaturalist website added machine le=
arning functionality to image observations, so now when an image is uploade=
d it will give you suggestions based on their machine learning image recogn=
ition model, in the video I sent
 before around the 3-minute mark, Alex Shepard says the model can ID to com=
mon ancestor (higher level taxon) with an accuracy of 93%. Each time an obs=
ervation is uploaded and vetted as &quot;research grade&quot;, that image i=
s added to the image analysis training library
 (used to train the model and make it more accurate down the line).&nbsp;</=
div>
<div><br>
</div>
<div>There's lots of other examples of Computer Science folks using Machine=
 Learning for this type of thing for citizen science (see Merlin like Rick =
mentioned), and also a bunch of more specific/academic purposes, though as =
far as I know, none are as accessible
 or ambitious as iNat. Here's some more info on the computer vision side of=
 things.&nbsp;<a href=3D"https://www.inaturalist.org/pages/computer_vision_=
demo">https://www.inaturalist.org/pages/computer_vision_demo</a></div>
<div><br>
</div>
<div><br>
</div>
<div>Cheers,</div>
<div>IM</div>
<div><br>
</div>
<div><br>
</div>
</div>
</div>
<br>
<div class=3D"x_gmail_quote">
<div dir=3D"ltr" class=3D"x_gmail_attr">On Sun, 17 Feb 2019 at 15:07, Fritz=
 McEvoy &lt;<a href=3D"mailto:fritzmcevoy@hotmail.com">fritzmcevoy@hotmail.=
com</a>&gt; wrote:<br>
</div>
<blockquote class=3D"x_gmail_quote" style=3D"margin:0px 0px 0px 0.8ex; bord=
er-left:1px solid rgb(204,204,204); padding-left:1ex">
<div dir=3D"ltr">
<div style=3D"font-family:Calibri,Helvetica,sans-serif; font-size:12pt">HI =
Ian,&nbsp;</div>
<div style=3D"font-family:Calibri,Helvetica,sans-serif; font-size:12pt">&nb=
sp; &nbsp; &nbsp;I've been on iNaturalist for a couple of years now and it =
is a great webpage. There are lots of sites like iNaturalist online now and=
 they are really helpful for having others help you
 identify a species.&nbsp;</div>
<div style=3D"font-family:Calibri,Helvetica,sans-serif; font-size:12pt">&nb=
sp; &nbsp; &nbsp;The thing I'm asking about is a program that would allow s=
omeone to upload a photo and/or maybe some data (size, colour, location and=
 whatever other ID info you have) of a fly, a leaf
 or a mouse etc. and have a AI computer program take that photo and/or data=
 and give you its guess as to Genus/species identification based on it's al=
gorithms - much like facial recognition and retina scans work today. All th=
e best.</div>
<div style=3D"font-family:Calibri,Helvetica,sans-serif; font-size:12pt">&nb=
sp; &nbsp; &nbsp; &nbsp; &nbsp; &nbsp; &nbsp; &nbsp; &nbsp; &nbsp; &nbsp; &=
nbsp; Fritz McEvoy</div>
<div id=3D"x_gmail-m_8501245089740205601appendonsend"></div>
<hr style=3D"display:inline-block; width:98%">
<div id=3D"x_gmail-m_8501245089740205601divRplyFwdMsg" dir=3D"ltr"><font fa=
ce=3D"Calibri, sans-serif" style=3D"font-size:11pt"><b>From:</b>
<a href=3D"mailto:naturens-owner@chebucto.ns.ca" target=3D"_blank">naturens=
-owner@chebucto.ns.ca</a> &lt;<a href=3D"mailto:naturens-owner@chebucto.ns.=
ca" target=3D"_blank">naturens-owner@chebucto.ns.ca</a>&gt; on behalf of Ia=
n Manning &lt;<a href=3D"mailto:ianmanning4@gmail.com" target=3D"_blank">ia=
nmanning4@gmail.com</a>&gt;<br>
<b>Sent:</b> February 17, 2019 1:43 PM<br>
<b>To:</b> <a href=3D"mailto:naturens@chebucto.ns.ca" target=3D"_blank">nat=
urens@chebucto.ns.ca</a><br>
<b>Subject:</b> Re: [NatureNS] AI and species ID?</font>
<div>&nbsp;</div>
</div>
<div>
<div dir=3D"ltr">Try uploading a photo to iNaturalist and seeing what it co=
mes up with. I think you'll be impressed.&nbsp;
<div><br>
</div>
<div>Cheers,</div>
<div>IM</div>
</div>
</div>
</div>
</blockquote>
</div>
<br clear=3D"all">
<div><br>
</div>
-- <br>
<div dir=3D"ltr" class=3D"x_gmail_signature">
<div dir=3D"ltr">We're mapping research in Southwest Nova Scotia.&nbsp;
<div>Link to your research project/publications <a href=3D"https://goo.gl/f=
orms/1JkoBRWIP3Kbvxl53" target=3D"_blank">
here</a>.</div>
<div><br>
</div>
</div>
</div>
</blockquote>
</div>
<br>
</div>
</div>
</div>
</div>
</body>
</html>

--_000_DM6PR18MB25062AFA0D05D1823819C023B7630DM6PR18MB2506namp_--

next message in archive
no next message in thread
previous message in archive
previous message in thread
Index of Subjects