Kia ora,

Welcome to the first newsletter of 2022. I hope you are all staying safe as Omicron runs through New Zealand. 

Once again, there is quite a lot to share in this newsletter:

• we have a new product: a thermal camera with a speaker attached to allow easy experimentation with audio lures
• there are some useful software updates making it easier to tag birds in audio recordings and improving the productivity of reviewing thermal videos from our cameras
• we have successfully implemented machine vision on our thermal cameras, allowing a host of innovative applications now the camera knows what it's seeing in real time
• the Cacophony team have continued their work on their AI powered trap.

In other news, my brother Grant, who is the founder of The Cacophony Project has been busy with a couple of other projects. He's published a book: Comparonomics about why life is better than you think and how to make it even better. Another one of his start-ups, Purepods, has announced their first North Island getaway.

Finally, with everything that is going on in the world, I thought I would share a thermal video I found the other day of a little blue penguin in Tāwharanui. It bought a smile to my face.

Ngā mihi nui,

Shaun Ryan

2040 Limited.

New product

Thermal camera with a speaker

Our latest product has actually been around as a prototype for a few years, but due to customer demand we have now made this available as a product anyone can buy. This is one of our thermal cameras with an additional connector for a waterproof speaker. It comes with software that lets you upload sounds and and set how often and at what volume you would like them to play. This lets you experiment with audio lures and the like. Read more about the thermal camera with a speaker. For details about the software, check out the manual.

Software updates

Better bird tagging and practical visits summary and more

We have released a new version of our bird tagging software that makes it easier to add bird tags and to see those that have been added, either automatically or manually. We hope that this will help us get enough data to expand the birds that we can automatically recognise (at the moment it is only morepork/ruru). Read more about the improved bird tagging update.

Another update earlier this year significantly improved our visits reporting for our thermal cameras.  You can now get a summary over a group of thermal cameras showing how many animal visits were detected each day. This significantly improves the productiveness of reviewing video footage. You can understand what happened overnight with a single glance and quickly drill down for more information.  Read more about the improved visits reporting.

AI prototyped on thermal camera

The camera knows what is in front of it

At the moment our thermal camera uploads the videos to the cloud where they are processed and the machine vision algorithms identify any animals. As it is this is really useful (check out the recent blog post from Cacophony about what a thermal camera will do for you). It can be made better: The Cacophony team have a prototype where the machine vision algorithms run on the camera and broadcast, via Bluetooth what that animal is. Other devices around the camera can then react to that signal. Example applications we have been talking to our customers about include:

• playing a barking sound to discourage animals from coming around a predator proof fence
• opening a door in a predator proof fence to allow penguins through to their nest
• playing a lure to entice an detected animal into a trap - eg play baby stoat sounds quietly when a stoat is detected
• enabling a leg hold trap when a target animal is detected. This could reduce the sort of bycatch that has been recently reported.

I'm sure there are more applications. If there is something you would like to try with this technology then please let us know. Here's a video demonstration of the machine vision working in real time on the thermal camera.

Trap update

The Cacophony team continue to work on the trap, which internally we are calling the Zepplin because the ramp is like a stairway to heaven.  As a reminder, this trap consists of a blind snap trap, with a live capture cage with a door at the end leading to a ramp that has an AT220 at the top.

The team is learning a lot as they test the trap. The animals consistently climb the ramp once they are trapped. The AT220 doesn't consistently kill the animals (this isn't a criticism of that trap because it's not being used in the way it was intended). We see this because the animals are in an enclosed area and can't wander away. The team continue to iterate on self resetting mechanism for a DOC 250 which has a lot more force and we hope will be more effective. That said, the trap has had it's first night with two kills.

We continue to capture footage of animals in the trap and are working to identify them automatically in real time so we can make a kill decision. This is using a small IR camera that is sitting on the inside of the trap.

2040 on social media

2040 is on Facebook, Twitter, Instagram and LinkedIn Follow us wherever you are to stay up to date.

Sign up to this newsletter

Have you been forwarded this newsletter?

Sign up here to receive this newsletter directly. We aim to send out one per month but often end up doing it less frequently because life gets in the way.

2040 Ltd is a social enterprise whose mission is to eradicate predators from NZ by manufacturing and selling open source technology developed by The Cacophony Project.