WILDLABS logoWildlabs.net : The conservation technology network

WILDLABS logoWildlabs.net : The conservation technology network

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.

event

WILDLABS Virtual Meetup: Low Cost, Open-Source Solutions

WILDLABS Team

18 March 2019 12:00am

The first event in Season Two of the WILDLABS Virtual Meetup Series was on Low-Cost, Open-Source Solutions. The recording is now available to watch, along with notes and recommended reading based on the discussion. In...

0
See full post
discussion

Conserving the Sumatran and Javan Rhino

claireoelrichs_siesfund_org

15 January 2019 7:39pm

Save Indonesian Endangered Species (SIES) will be building a machine to identify individual Indonesian rhino.  This will be challenging due to low data numbers.... Sumatran...

5 0
Colin Kingen

23 January 2019 10:12pm

Hello Claire, 

Engineer at Wild Me here. We would love to start a conversation about a Wildbook for rhinos. 

Lets talk about citizen science and computer vision for identification possibilities. I'm curious about your current data set and the identification tools you are using as a starting point. I'm happy to talk here, or you can email our team at services@wildme.org.

Ed Miller

24 February 2019 1:28am

Hi Claire,

At the BearID Project, we are working on a similar problem for brown bears. We are currently using machine learning methods developed for human facial recognition (like Google FaceNet). We got some ok initial results, but now we are running up against small data issues. The method for human faces were trained with millions of images of hundreds of thousandes on individuals. We have a few thousand images of about a hundred individuals. We plan to investigate other methods in the future.

It will be great to keep in touch to see what methods you will be using.

Ed

Ed Miller

24 February 2019 1:38am

Hi Colin,

At the BearID Project, we are working on a similar problem for brown bears. We are currently using machine learning methods developed for human facial recognition (like Google FaceNet). We got some ok initial results, but now we are running up against small data issues. The method for human faces were trained with millions of images of hundreds of thousandes on individuals. We have a few thousand images of about a hundred individuals. We plan to investigate other methods in the future.

The last time I talked to WildMe, the identification algorithms were based on matching unique patterns. We didn't think this would be directly applicable for brown bears as they don't have a lot of clearly identifiable markings. Have you developed other identification algorithms?

Ed

See full post
discussion

Responsible AI for Conservation?

Ollie Wearn

11 February 2019 6:22pm

AI is booming in conservation, with almost daily news articles on how it will solve problem X with algorithm Y. Indeed, there's huge potential here. But, as the old adage goes...

4 0
Ollie Wearn

21 February 2019 8:17am

Hi Jaishanker

Absolutely - the overlap between image-based and sound-based analyses is increasing, and consistent terminology will no doubt help us share info. 

Are you using ML in SODA for automated identification of sounds? If so, how are you determining if a given classifier is performing well? 

Thanks
Ollie

Jaishanker Nair

21 February 2019 11:08am

Hello Ollie,

SODA is a recently launched suite. It is in the development phase. We have with us call libraries with multiple (40+) calls for 10- 12 species of birds. A research scholar is on the job for classifying at the species level.

Our interest is equally on separating the different sonic components (as stated in https://www.wildlabs.net/community/thread/666). It is different from the link shared in my previous reply. This is where I see the confluence of objectives.

As a TEAM, I believe, we can address the individual objectives faster.

regards

jaishanker

 

Laura James

21 February 2019 9:13pm

Hi Ollie, 

Great article, thank you!  I mostly work with responsible AI in other contexts, at Doteveryone.org.uk and the Trust & Technology Initiative at the University of Cambridge, so don't have much to offer here, although I am very interested in the topic.  I appreciate your point that many of the consumer data issues highlighted in the 'popular' responsible AI discourse aren't relevant to conservation (some of us have been gathering 'responsible tech' / 'ethical tech' resources in a shared doc, and there's essentially nothing there for conservation specifically - https://docs.google.com/document/d/1SN6hYeKe3eRK6x9D0Sr7GpCA4nirpyo3u68xG1A6NDs/edit ). However there might be some links with humanitarian data practices, which are touched on by the Responsible Data folks at https://responsibledata.io and https://www.fabriders.net/data-literacy-consortium/ or in this recent article https://asecondmouse.wordpress.com/2019/02/20/instability-forecasting-models-seven-ethical-considerations/

Best, 

Laura

See full post
funding

The Ecosulis Rewilding Tech Challenge

Cain Blythe

14 January 2019 12:00am

With the aim of advancing rewildling-related technology in the UK and introducing new talent and ideas into the field of rewildling, Ecosulis is thrilled to announce the launch of their first ever Rewilding Tech...

0
See full post
event

WILDLABS Virtual Meetup: Big Data in Conservation

WILDLABS Team

27 November 2018 12:00am

The third and final event in Season One of the WILDLABS Virtual Meetup Series was on Big Data in Conservation. The recording is now available to watch, along with notes that highlight the key takeaways from the talks...

0
See full post
event

Tusk Conservation Lecture 2018: Ted Schmitt

Tusk

15 October 2018 12:00am

Tusk are delighted to announce that their Tusk Conservation Lecture 2018 speaker will be Ted Schmitt of Vulcan Inc, Paul G Allen Philanthropies. He will speak about advancing Innovation in conservation, a talk that will...

0
See full post
event

Ocean Hack: San Francisco, 10-11th September, 2018

One Ocean Collab

20 August 2018 12:00am

A 48 hr pop up innovation lab for the ocean, bringing together a mix of designers, strategists, technologists, engineers, scientists, marine conservationists, educators, artists and buisness talent to co-create...

0
See full post
article

Thermal Sensor Project Update: Testing with live animals at the San Diego Zoo

Arribada Initiative

27 July 2018 12:00am

The winners of our Human Wildlife Conflict Tech Challenge are offering regular updates throughout the year to chronicle their failures, successes and what they learn along the way as they develop their solutions. In...

1 0
Lars Holst Hansen

31 March 2023 2:15pm

Hi Anne!Interesting with the vaccum seal issue!When looking at the sensor comparison images, it seems like there is not only a difference in noise (or SNR). The Heimann 80x64 also...
See full post
article

FIT Cheetahs

Larissa Slaney

4 December 2017 12:00am

A new research project is looking to investigate whether technology combined with the ancient skills and knowledge of Namibian trackers can help save cheetahs from extinction. Called FIT Cheetahs, the research project...

0
See full post
discussion

Automated video count of migratory birds

Steffen Oppel

22 April 2016 8:38am

We are wondering whether anybody is aware of a free/open-source system to extract and count moving signals from a videostream, with the ultimate goal to apply such a system to...

1 0
Stephanie O'Donnell

30 October 2017 11:46am

Hey Steffen, 

I know you've had a student working on this challenge for the past year - how is this project progressing? If you (or your student) have a moment, it would be great to hear an update.

@mmckown shared an in depth write up of one their projects that I thought might be relevant, as it seemed they were tackling something similar to what you are looking into? His team at Conservation Metrics (which presumably included @kleinsound) partnered with Microsoft to automate counts of Red-legged Kittiwakes with ML. I know it's not the exactly the same problem you're looking into, however the post covers their end-to-end flow for object detection, so might have some useful ideas/approaches that may have relevance for your work.

Bird Detection with Azure ML Workbench

Introduction

Estimation of population trends, detection of rare species, and impact assessments are important tasks for biologists. Recently, our team had the pleasure of working with Conservation Metrics, a services provider for automated wildlife monitoring, on a project to identify red-legged kittiwakes in photos from game cameras. Our work included labeling data, model training on the Azure Machine Learning Workbench platform using Microsoft Cognitive Toolkit (CNTK) and Tensorflow, and deploying a prediction web service.

In this code story, we’ll discuss different aspects of our solution, including:

Steph 

 

 

See full post
article

HWC Tech Challenge Update: Meet the Judges

WILDLABS Team

20 October 2017 12:00am

Our panel of international experts has been hard at work reviewing the 47 proposals we recieved for innovative technological tools to address human wildlife conflict. The panelists have systematically been assessing the...

0
See full post
discussion

[ARCHIVED] Fish identification computer vision competition

Kate Wing

26 September 2017 8:43pm

Our contest to develop tools that can count, measure, and identify fish from commercial fishing video runs through October 30. Please pass this on to anyone you know who might be...

1 0
Stephanie O'Donnell

29 September 2017 1:00pm

Hi Kate, 

It's really exciting to hear that you've now launched the challenge, congratulations on getting to this point! It's going to be interesting to see what solutions come out of the challenge - please do keep us updated as it progresses if you have time. The challenge is focused on the New England fishery - are you envisaging that this is an approach you can take to scale and eventually extend to other fisheries? 

To add a bit more information for anyone interested, there's actually $50,000 of prizes attached to this challenge:

Place Prize Amount 1st $20,000 2nd $15,000 3rd $10,000 4th $3,000

There is also a wildcard prize:

We're also looking for innovative approaches to solving this fishy problem, even if they don't score in our Top 4. If you want to be eligible for our $2,000 Judges' Choice Award, submit your code on the Submit Report page (available once you've signed up) by the competition end date for review. Our judges will be looking for inventive, novel solutions that can be incorporated into video review programs, so share your most fin-tastic ideas. 

Finally, if you're curious to find out more about what led to the challenge, Kate actually wrote a piece called 'Machine learning, meet the ocean' that we published in the resources area a few months ago. Do have a read!

Cheers,

Steph 

 

See full post