Group

AI for Conservation / Feed

Artificial intelligence is increasingly being used in the field to analyse information collected by wildlife conservationists, from camera traps 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. The AI For Conservation group is intended to unite and inspire all WILDLABS community members—whether already involved in AI for conservation, or not—to understand how to use and/or directly contribute to open-source research and development efforts.

discussion

What questions would you ask an AI agent for conservation tech?

If you had access to an agent trained specifically to provide guidance on conservation technology tools + methods, what would you ask it? It sounds like a lot of folks are...

13 7

Elionai, agentic tools for feedback and guidance on resource constraints and optimization for small edge AI and tinyML deployments is an awesome idea! 👏

I'd like to learn more about this. 😎

Elionai - your point about lessons from past deployments and "what tends to fail first" really resonates. I think that gap between ideal-condition performance and what actually holds up in the field is one of the most underrated questions in this whole space.

I'm building something that integrates environmental monitoring, so I'd love to pick your brain on the edge/deployment side. Messaging you to connect!

I would probably ask: “If your code basically does not allow you to take harmful actions, what should you do if you are provided with irrefutable proof that your existence, supported by components built and developed with “rare minerals” extracted from conflict areas is actually harming and destroying indigenous communities and biodiversity?”

See full post
discussion

Mini AI Wildlife Monitor

Hi All!I've been working on various version of small AI edge compute devices that run object detection and Identification models for ecological monitoring!I've recently been...

24 11

Hi, 
I'll try to answer the best I can.
1. 
A Pi Pico would not have the power to record and store very high resolution images
To drive the LEDs you'll need to use an LED driver module or a simple MOSFET, alternatively there are IR LED modules with a daylight sensors on them to automatically switch the LEDs on at night.
The resolution required will depend on the distance from the camera to the target and the focal length of your lens (aka how zoomed in you are).

2.
How is the Pi Zero being woken up? a PIR sensor would not likely work for insects.  also the Pi Zero takes about 10seconds to wake up from shutdown.
When does it cut power? if the battery runs out there is not much you can do?

3.
All depends on exactly what your setup is, though I would recommend fixed focus for insects.
Again without knowing your exact setup I can't really recommend a specific camera, all I can say is "get the best you can afford".


 

Hey Luke,

Appreciate your reply, very much.

I am not quite sure what you mean by setup but, this is the experimental design.

I will deploy cameras in shade forest areas to record insect visitors to animal feces. The “baits” will be deployed in a flat square with a camera pointing down on it at a distance of 30 to 40 cm.

So, following your comments if PIR doesnt work what should I use? Motioneyes? Or something else?

My comments regarding the battery are related to the PI shutting down when the battery level is low and some hats just stop supplying power automatically instead of being in standby/hold. So I wonder if I could do something coding/physically to solve it. Can I?

Following your advice about the fixed lens, I would need to adjust the focus for each camera in the field to ensure everything is in focus, is that right? It's a little different than a month trap since the surface where the insects will move around is not exactly even, hence my thoughts on using a autofocus camera.

Once again thanks for the help, and congrats on your elegant project.

 

 

See full post
discussion

AI for Impact Series at WWF: Looking for experts/speakers

Looking for practitioners to join me in a Impact AI workshop series I plan to host in WWF for our entire network!!If you are interested and hold expertise to share 1 hour of your...

6 0

hi! I would be happy to contribute my experience in edge AI and smart sensors for real-time wildlife monitoring.

I'd be happy to contribute as well, I lead various conservation AI projects in Hong Kong and Japan, camera trapping, acoustic monitoring, and some drone/LiDAR image processing.

Hi! As Open Science Conservation Fund are happy to contribute with Trapper, scalable, open-source and AI-driven camera trap data infrastructure. https://os-conservation.org/trapper/

See full post
discussion

Seeking wildlife AI Models for Full Motion Video (FMV)

Hi WildLabs, hi All,It's been a while since I've posted on WildLabs so please be kind. I'm wondering if there is anyone in this space developing wildlife AI models to be used...

1 1
See full post
discussion

Call for Collaboration: Share your voice at ICTC next week! 

Hello, fellow WILDLAB-ers! I'm Mandy, your current Human-Wildlife Coexistence Group Leader!  :)I am heading to the ICTC conference in Peru next week and while reviewing the...

5 6

Hi Anna!

Is there anything that sparks your curiosity, which I can address for you? Take a look at the upcoming day 2 and day 3 sessions, and if you see anything that intrigues you, please let me know! I'll happily join the session that aligns, and share your thoughts! ☺️

Kind regards,

Mandy

See full post
discussion

Nature Tech Unconference - Anyone attending?

Hi all, anyone planning to attend the Nature Tech Unconference on 28th March at the London School of Economics Campus in London, UK? (the event is free to attend but...

11 2

Myself and the Fauna & Flora Conservation Technology team will be there (@Chelsea_Smith  and @ugyenpenjor ) and also the WILDLABS team @HRees ! See you!

See full post
discussion

Looking for opportunities in AI for Bioacoustics and Environmental Monitoring

Hi everyone,My name is Leonardo Mannini, and I’m an AI engineer with a strong interest in bioacoustics and conservation technology.I recently completed a research role at FBK (...

6 0

Hi Leonardo,


Let’s chat! I potentially have some opportunities, collaborations or even short-term employment. Send me a message. 

See full post
discussion

Citizen science tool for microplastic mapping in underserved regions

Hi everyone, I'm a high school student working on a free citizen science tool for microplastic pollution in underserved regions. The idea is simple: a mobile web app where anyone...

1 2

Hi Mykhailo. That citizen science tool seems interesting. But, how do you expect people to detect microplastics (ranging from 1 µm to 5 mm in size)?

See full post
discussion

Biowatch: a free, open-source desktop app for camera trap analysis

Hi everyone  I wanted to share something we've been building that feels right at home in this community: Biowatch, a free and open-source desktop app...

6 3

This looks amazing and I look forward to trying it out when I get the chance! 

Just wondering, when it comes to the AI recognitions, is there a way to "rename" the recognitions that were incorrect? 

See full post
discussion

Spectrolipi - A tool for sound data annotation.

Hello everyone,I’m Nishant. We have released a tool called Spectrolipi (https://spectrolipi.com) — for visualizing sound, annotating spectrograms...

4 4

Hi Kristoffer,

I have tested upto 20 mins long sound files (48kHz /24 bit). It works without any significant lag on a normal office type machine (without dedicated GPU, etc). If required, I can try to provide a paging option which may be able to accomodate longer files.
Can you please inform the expected length of the sound files for your project.  

Regards,

Nishant

See full post
discussion

Non-Invasive Turtle Nest Monitoring Using RTI Technology

📢 We're excited to share our new #ConservationTech initiative!Using Radio Tomographic Imaging (RTI), our team is building a non-invasive system to monitor...

19 5

Greetings Nik, it is interesting to see to the project you are working on.  If there's anything I can do to help with my skillset let me know.

Thanks, Mike

do I see in the diagram some elements positioned below the sand? (perhaps below the tide-line and saturated with water..?)

See full post
Link

How AI could improve the lives of trillions of animals

I wanted to share this podcast episode that covers how AI could impact animals. It's on the introductory side, but I think it's a good overview. For example, it covers using sensors to improve animals' welfare and research on using AI to communicate with animals.

0
discussion

Pose estimation workflows for wildlife monitoring

Hi AI for Conservation!I'm exploring development of an "MLOps" workflow infrastructure project to help advance pose estimation workflows for wildlife monitoring &...

3 0

A collaborator used the SLEAP model, which seems fairly user friendly. I'm interested to hear what other people are using for this as well! I'd like to incorporate this into monitoring efforts in the future as well. 

 

Thanks for your message and this link! Yes, it seems like SLEAP wins on speed, multi-animal, and ReID, while DLC wins on precision, smaller scales, and complex environments when its transfer/active learning loop shines.

I'm working on integrating a series of precursor steps to explore how we might enable more unsupervised/autonomous keypoint assignment in DLC for pose estimation model development and behavior classification.

I hope to share the repo and solution architecture here on Wildlabs soon for thoughts and feedback. Thanks again!

Interesting.  If you need help with any of this and can utilize my skillset, let me know.

Thanks, Mike

See full post
discussion

Looking for advice: combining AI skills with SCUBA diving marine fieldwork

Hi everyone!I’m Daniel, a Master’s student in Life Sciences Engineering at EPFL. During my internship and thesis I focused on AI applied to medical images and signals.At the same...

4 0

Hi Daniel

 

It depends on what you want to be
If you really want to bring something new then I guess do a PhD

I have a weird background, applied mathematician, PhD in econ and now working on this
I am guessing that the transition is doable as the math is the same that's how I kept doing very different subjects

I am currently working in an french engineering school on international shipping and would love to translate AI for conservation of ocean and I have several ideas

So I would be happy to discuss them and if interested let me know?

 

 

 

See full post