Hacker News new | past | comments | ask | show | jobs | submit login

Any interest in sharing the plane dataset you used to train Azure Custom Vision with others? Happy to host it on Roboflow Universe along with an API to a trained model others can hit to detect planes in images.

We’ve got one detecting planes on the ground but not yet in the air! https://universe.roboflow.com/gdit/aerial-airport




Definitely - Let me build a little better dataset. I need to augment a bit because it was false detecting on some of the contrails.

I built some scripts to also start building a model that can classify the different aircraft models. One of the interesting things with this, is that you get really accurate labels... probably better than what a human could do for images of this size.

https://github.com/IQTLabs/SkyScan/tree/main/ml-model


Did they ever indicate using a model like that? Just plug in an ADS-B receiver and they'll broadcast their tail number and location to you. Everything else can be looked up.


Yeah the crop is done after uploading a photo taken towards the general direction of the plane:

> The Function uses a model trained in Azure Custom Vision to detect if there is an airplane in the photo and where it is


Oh that's interesting. It doesn't even have to be that good at plane identification, just "what's the main focus of this picture".




Consider applying for YC's Spring batch! Applications are open till Feb 11.

Guidelines | FAQ | Lists | API | Security | Legal | Apply to YC | Contact

Search: