Group

AI for Conservation / Feed

Artificial intelligence is increasingly being used in the field to analyse information collected by wildlife conservationists, from camera trap and satellite images to audio recordings. AI can learn how to identify which photos out of thousands contain rare species; or pinpoint an animal call out of hours of field recordings - hugely reducing the manual labour required to collect vital conservation data.

discussion

Looking for animal emotions data for ML

Hi. I'm hoping this community is able to help as there are probably many pet parents here: We're researching animal identification and emotion recognition with ML...

2 0

There is a TON of research on ML algorithms for all this!

For facial rec-

BearID project - http://bearresearch.org/ (individual facial rec for grizzlies)

AnimalFACS (facial action coding system) - http://www.chimpfacs.com/ (the model started with chimps but they now have versions for orangutans, gibbons, macaques, dogs, cats, and horses)

A lot of the gesture/facial expression literature is focused on primates, particularly apes, so if you do a Google Scholar search on that, you'll loads of papers that have looked into this as well.

For vocals - 

Carlson, N. V., Kelly, E. M., & Couzin, I. (2020). Individual vocal recognition across taxa: a review of the literature and a look into the future. Philosophical Transactions of the Royal Society B, 375(1802), 20190479.

Turesson, H. K., Ribeiro, S., Pereira, D. R., Papa, J. P., & de Albuquerque, V. H. C. (2016). Machine learning algorithms for automatic classification of marmoset vocalizations. PloS one, 11(9), e0163041.

Wijers, M., Trethowan, P., Du Preez, B., Chamaillé-Jammes, S., Loveridge, A. J., Macdonald, D. W., & Markham, A. (2020). Vocal discrimination of African lions and its potential for collar-free tracking. Bioacoustics, 1-19.

Clink, D. J., Crofoot, M. C., & Marshall, A. J. (2019). Application of a semi-automated vocal fingerprinting approach to monitor Bornean gibbon females in an experimentally fragmented landscape in Sabah, Malaysia. Bioacoustics, 28(3), 193-209.

Spillmann, B., van Schaik, C. P., Setia, T. M., & Sadjadi, S. O. (2017). Who shall I say is calling? Validation of a caller recognition procedure in Bornean flanged male orangutan (Pongo pygmaeus wurmbii) long calls. Bioacoustics, 26(2), 109-120.

Hantke, S., Cummins, N., & Schuller, B. (2018, April). What is my dog trying to tell me? The automatic recognition of the context and perceived emotion of dog barks. In 2018 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP) (pp. 5134-5138). IEEE.

Totakura, V., Janmanchi, M. K., Rajesh, D., & Hussan, M. T. (2020). Prediction of Animal Vocal Emotions Using Convolutional Neural Network. International Journal of Scientific & Technology Research, 9(2), 6007-6011.

Other - 

Neethirajan, S., Reimert, I., & Kemp, B. (2021). Measuring Farm Animal Emotions—Sensor-Based Approaches. Sensors, 21(2), 553.

Hong, W., Kennedy, A., Burgos-Artizzu, X. P., Zelikowsky, M., Navonne, S. G., Perona, P., & Anderson, D. J. (2015). Automated measurement of mouse social behaviors using depth sensing, video tracking, and machine learning. Proceedings of the National Academy of Sciences, 112(38), E5351-E5360.

 

Thank you VERY much for that, Carly. I really appreciate it. I have only heard of the bear research project and was not aware of the others. This is very helpful.

FYI: Most studies I am aware of use ML with facial recognition software. I am interested to find out if there are recognition software/research for facial-gesture-voice-speech as this will be more accurate to read animals that have less elastic facial expressions than primates or humans.

See full post
event

Event: tinyML for Good: Conservation & Climate

Edge Impulse
Join the tinyML Summit 2021 tomorrow, March 23rd at 12 PM PT for the panel discussion "tinyML for Good: Conservation & Climate," featuring Talia Speaker from WILDLABS. Register for this session here, and view the...

0
See full post
article

Building Experts Into AI

Whale Seeker
In this article from Whale Seeker, you'll read about how human expertise plays a role in building strong AI algorithms and achieving accurate wildlife identification results, as well as the ethics of using AI in these...

0
See full post
article

Kaggle Competition: iWildcam 2021 - FGVC8

CVPR
This year's iWildCam competition is now live on kaggle. Go beyond just classifying species or detecting animals - this year the challenge focuses on counting how many individuals of each species are seen in a burst of...

0
See full post
funding

Funding Opportunity: COVID-19 Science Fund

National Geographic
National Geographic is offering funding up to up to $50,000 for conservationists conducting research on how the pandemic has impacted wildlife and conservation work.  If you are interested in researching aspects of the...

0
See full post
article

Resource: WildID

WildID
WildID is excited to share their new camera trap processing and detection tools with WILDLABS! Using machine learning to identify Southern African wildlife species in large quantities of camera trap data, WildID's tool...

0
See full post
discussion

Looking help with camera trapping for Jaguars: Software for species ID and database building

Hi! I'm looking for help with the development of a software for the analysis of thousands of photos from camera traps. We have more than 200 cameras deployed in field and...

5 0

Hi Carmina,

I definitely get not wanting to create another copy of your data, I realized this about Camelot just a little late, and we did have to buy a 4 TB external hard drive. The unique names that Camelot assigns to its copies of the data are also not friendly if you want to move things around.

I mentioned that Wild.ID is being more and more integrated with Wildlife Insights, but it can still be used as a standalone software. It has the same format as Camelot, but I think it doesn't create and additional copy (I might be wrong about this, though). You should take a look at this, it might be what you need.

As for detection, Camelot puts together batches of photos, uploads them to Microsoft's servers, and then gets the resulting database back. The model they use, however, is available and you can run it on your own computer. The AI model that it uses is called Megadetector. You used to have to run it from the command line, but recently Petar Gyurov posted this GUI that makes using Megadetector soooo much easier. You just tell the software where your photos are, and let it run. You can decide whether you trust it enough to sort your photos automatically, or check yourself. In most conditions it performs very well, you may have issues when you can only see parts of an animal behind some vegetation. The check would take less time than going through the raw pictures, since it marks boxes where it found things. You still have to identify the pictures after they're sorted. Megadetector will work on any computer, but it performs much faster if it has an Nvidia graphical card. However, setting this up is a little tricky.

Finally, the output from Megadetector can be integrated into the workflow of Timelapse, another software. I've never used this one so I can't really say how well it works.

Integrating everything can be a little intimidating, so feel free to ask if you need guidance for any of these steps.

Best,

Juan

 

Hi.

Thanks a lot for your explanation. I decided to try megadetector but I'm dumb with computers and of course I don't have a Nvidia graphical card, can you help me?

If it is better for you to talk directly, you can write to [email protected]

I'm so happy I can finally see the light with so many photos :)

Thanks again

Carmina

See full post
discussion

Transparency, AI, and Machine Learning Conservation Impacts

Hi all, I'm writing up a short article about the rewards and risks of using big data and AI for transparency in conservation. I'm pointing to sites like...

2 0

Hello,

Just spit-balling some ideas, but you might check out WildMe/WildBook, Widlife Insights, the ICARUS/Movebank project, Arbimon, Norway's new imaging initiative (https://www.planet.com/nicfi/), Hack the Poacher, Vulcan's EarthRanger, SmartParks, the SMART app/platform, the TEAM camera trap network?

Not sure if any of this is what you're looking for but just some things that came to mind! 

Thanks Carly,

These are great ideas. The movebank recommendation reminds me of some work from Save the Elephants that used its transparency to exonerate an elephant that had been accused of eating up some crops. I didn't know about Hack the Poacher - I'll check that out. I'm familiar with the others, but will dig deeper to understand the transparency angle.  I work on Wildlife Insights, so know that one well. TEAM has essentially morphed into Wildlife Insights as I understand things. 

Thanks again!

Best,

Thau

See full post
discussion

Windows explorer video thumbnails

Does anyone know a way of making Windows explorer video thumnails show a frame from early in the video, when the animal that triggered the camera is still in view ? It would then...

1 0

This works very well, and can remove the movie strip icons from the edges of the thumbnails, which obscure animals just coming into the frame. https://icaros.en.softonic.com/

See full post
article

Intro to Embedded Machine Learning on Coursera

Edge Impulse
Edge Impulse is proud to announce their brand-new online course, Introduction to Embedded Machine Learning. Hosted on Coursera, this professional training course will provide beginners with the tools to started with...

0
See full post
discussion

Tech Tutors: How do How do I launch machine learning projects using MLOps?

Hi everyone, This week we are pleased to welcome back EdgeImpulse's Daniel Situnayake. Dan will be building on his first tutorial, where he walked us through training...

1 0

Dan's comments about the need for technologists and conservationists to manage and share (properly annotated) data struck a chord with me, it was right at the end of the presentation.

I fired off a point into the chat "could your old background data be my background data?" or something similar, and it got me thinking...

(Firstly, sorry for clouding this issue with my simultaneous "Ian Tuna" joke)

In the context of using AI as described in EdgeImpulse example, lets say - audio.

Lets say Group A are trying to detect the roar of lions, and Group B are trying to detect the grunt of wild pigs, and lets imagine that is in roughly the same area of Africa.  I mean they could both send out teams to capture the sound of their target animal, and the sound of NOT their target animal, e.g everything else.

Well if "serengetti sounds" was a known audio track, then all you need is the unique sound of the beast in question, am I right?

See full post
discussion

Seeking interviewees - AI in Conservation & nonprofits

Hello, My name is Alexander Lam. I am an undergraduate honors student investigating the use of artificial intelligence/machine learning by conservation nonprofit organizations...

9 0

Hi @Alex.L , I would suggest the same as for @mattzig at https://www.wildlabs.net/community/thread/1046#post-4348 :

I could try to connect you with some of the people and orgs we have worked with. If you go through https://www.vizzuality.com/project/ and identify the projects you would like to be look into, I can then try to connect you with the best person for each project.

And I would be very interested in learning more about your research group's work, besides your current research question, if possible!

Happy New Year! Luisa

See full post
discussion

Tech Tutors: How do I start a Tech4Wildlife collaboration?

Hi Wildlabbers, Ed Miller and Melanie Clapham tackled the...

2 0

One barrier that was discussed in the presentation was the "risk" or uncertainty of a payoff when talking to potential funders. It seems to me that the "risk" factor of collaborations like this, for funders, would be analogous to the investment and payoff when developing a genomics program. Relative to traditional field research, it takes a much greater investment in time, with different expertise, to create a reference genome for a species of interest and then pull SNPs, etc. Nevertheless, funders seem to recognize the potential for genomic research and genomic tools. Do you think that greater openness to “risk” in conservation genomics results from closer intellectual proximity (i.e., molecular biology vs organismal biology), or to conservation genomics being a more mature field (i.e., more examples of success), or to other factors? If intellectual proximity is a key factor, then conservation tech collaborations will continue to face this barrier, but if lack of successful examples are a key factor, then perhaps this barrier will become less of an issue for conservation tech in the future. Thoughts?

Hi. Just watched the Youtube version (I'm in the unable to watch live hemisphere) and wanted to say that was a great talk. I wish it could have gone for another hour. I liked hearing about the issues and concerns from Melanie on the biology/ecology side. Also could totally relate to what Ed was talking about from the tech side about how it's difficult to have long term collaborations since there isn't really a conservation technology career path at the moment. 
I think one of the things we're trying to do with courses like Build Your Own Datalogger is to get people involved in wildlife/conservation side more comfortable with the tech. The goal is to hopefully build a shared repository of technical knowledge that people in the wildlife community can contribute to and benefit from. 

Akiba

See full post
article

Protecting Wildlife with Machine Learning

Hack the Poacher
Last year, Tim van Deursen and Thijs Suijten shared their new "Hack the Poacher" system with us, presenting a unique way to detect poachers in real-time within protected national parks. Read on to learn about their next...

0
See full post
event

How do I launch machine learning projects using MLOps?

WILDLABS Team
Our eighth Tech Tutors Season 2 episode featured Edge Impulse's Dan Situnayake who tackled the question: How do I launch machine learning projects using MLOps? Watch it on the WILDLABS Youtube Channel! You can also ask...

0
See full post
event

How do I start a Tech4Wildlife collaboration?

WILDLABS Team
Our seventh Tech Tutors are BearID's Melanie Clapham and Ed Miller, who tackled the question: How do I start a Tech4Wildlife collaboration? Lessons learned from developing facial recognition for bears. Watch it on the...

0
See full post
discussion

X-Prize for Rainforests

https://rainforest.xprize.org/prizes/rainforest X-Prize have announced a $10m prize "to survey the most biodiversity in at least three stories of a rainforest (emergent,...

29 0

I think if you added it to your calendar through the registration button on the Rainforest XPrize site, it added the link to join to your calendar.

Hi everyone!

I only just came across this thread and the Rainforest XPrize. I would love to be a part of the team if it's not too late. I filled out the Google Sheet form with my experience, but I also wanted to post a note in the form. I have experience managing conservation tech projects (with some members of this forum!), testing tech in the field, and conducting biodiversity surveys in the Amazon. I read through the Google Docs notes from previous meetings and you all have some interesting ideas. I would love the be a part of future planning meetings if you'll have me! 

See full post
discussion

Tech Tutors: How do I use pattern matching analysis to label acoustic data with RFCxArbimon?

Hi Wildlabbers, Tech Tech Tutors Season 2 continues with our third epsiode, featuring Rainforest Connection's Zephyr Gold & Marconi Campos Cerqueira‬, who tackled...

3 0

Thanks everybody for joining today's session! I've attached the article here that describes Pattern Matching in greater detail to respond to some of the questions from today's session. 
Also, here is a link to our Privacy Policy and our Terms of Service for the individual who had questions about GDPR. Feel free to contact us at [email protected] with any other questions about the platform or Pattern Matching in general! 

Zephyr and Marconi, thanks for the great talk and your hard work that made this platform available to the public!! I'm experimenting with the pattern matching function to create some training data but have issues trying to get the "Jobs'' to run. I only have 18 files for each of these tests, but only 1 processed to 11.1% while two others stay at 0%. The internet connection was briefly interrupted when I created the first job and was fine later on. I can only Hide, not delete or re-run these jobs, so I'm not sure what to do. Any suggestions would be greatly appreciated. Thanks so much!

Hi @pmnguyen1224 , thanks for reaching out and checking out the system! We would love to help ensure that you're able to get pattern matching to work for you! I do have a few questions for you so we can replicate issue you are experiencing. Would you mind emailing us a couple screenshots and your original message above at [email protected] to get started? Or feel free to send us a chat if that's easier, by clicking on the chat icon on https://arbimon.rfcx.org/. Looking forward to supporting you moving forward!

See full post
article

British Antarctic Survey: Monitoring Whales From Space

British Antarctic Survey
Scientists have found that studying high-resolution images of whales from space is a feasible way to estimate their populations. A team led by British Antarctic Survey (BAS) compared satellite images to data collected...

0
See full post
discussion

Realtime video analysis using ML on Raspberry PI

HI, I am looking into using ML to identify a flock of birds from a video camera feed, either color or grayscale, using OpenCV and Python on a Raspberry Pi? Has anyone attempted...

1 0

Hi Andrew,

Yo need to train a lightweight DNN model for bird flocks which can then be deployed on Raspberry Pi. For initial starters, you can look into the below tutorial:

 https://www.pyimagesearch.com/2017/10/16/raspberry-pi-deep-learning-object-detection-with-opencv/

Best

Piyush

See full post
article

FIS Call for Expressions of Interest (closed)

Fisheries Innovation Scotland
WILDLABS community members are invited to submit an Expression of Interest to Fisheries Innovation Scotland (FIS) to participate in two research projects involving conservation tech's role in the future of sustainable...

0
See full post
event

Making the Most of Tech Tutors Season 2!

WILDLABS Team
WILDLABS is celebrating its five year anniversary! Throughout the rest of 2020, we'll be sharing articles, community features, and case studies showcasing the incredible projects, collaborations, and successes that this...

0
See full post
event

WILDLABS Tech Tutors: Season Two

WILDLABS Team
The WILDLABS Tech Tutors are back! Starting this December, join us for our second season and get even more answers to your biggest "how do I do that?" questions of conservation tech. Whether you're a #tech4wildlife...

0
See full post
funding

Kaggle Competition: Species Audio Detection

Rainforest Connection
Hey Acoustic Monitoring and AI for Conservation community members - don't miss Rainforest Connection's Species Auto Detection Kaggle competition, open for to competitors and teams now! Participants will have the chance...

0
See full post