Hacker News Reader

Comment by mycowerk | Hacker News Reader

BirdNET and Pl@ntNet have to be my favorite uses of AI to date. A true use of technology for the greater good.

I've actually been thinking about automated bird recognition for a while now, I live underneath a flight corridor for migrating common cranes (near Berlin). I'd love to be able to one day track their migration across the continent in real time using data from crowdsourced base stations.

I wonder if migrating birds could be identified in flight using optical/radar/audio. If anybody else has had similar ideas I'd really love to chat on this topic.

View on HN

The iNaturalist app is also quite amazing for identification of all kinds of organisms by image. The also have the "Seek" app which identifies species even offline.

The Seek app is one of those things that you can show people and they think it is magic, even in this day and age. I got my mom and my grandma using it!

I use both apps, they’re pretty amazing. I typically use Seek first and if it can’t identify the image it lets you upload it to iNaturalist seamlessly for manual identification. I believe then those identifications are then used to further train the model.

Many enthusiasts and scientists are recording and identifying birds from nocturnal audio recordings. Some references:

- https://www.facebook.com/groups/NocturnalFlightCalls

- https://github.com/HaroldMills/Vesper

- http://oldbird.org/pubs/fcmb/start.htm

- https://nightmigrants.com/main/page_home.html

Very cool, thanks for the resources. Awesome rabbit hole to go down!

I've been toying around with photogrammetrically matching in-flight bird silhouettes in order to have a plantnetlike bird classifier. Very much in the data acquisition stage but I'm seeing moderate success in clustering images by bird shape so far. In between other photogrammetry work I'm starting to think about how to use colour information and reading up on how to select and train a ML model for this problem.

I don't know if you've ever photographed birds, but I suspect this would be difficult because shots that are detailed enough to make out enough detail would be tight in, and shots that covered a large patch of sky would not be detailed enough.

Just speculation though, I'm sure with enough resources these are surmountable issues... I'm just not sure what quantity of resources that is.

Yeh this was my initial concern with optical silhouette classification. The altitude and targets are quite small (and rapidly moving). I've thought about using optical flow analysis to track wingbeat frequency etc as I feel this might be a little simpler to acquire.

On wingbeat frequency I'm very curious to play around with mmWave doppler radar (there's some RF on chip stuff around), emitting a relatively isotropic signal might allow for a pretty wide field of view and I imagine you'd get a pretty reasonable wingbeat signal.

Reply on news.ycombinator.com