This forum uses cookies
This forum makes use of cookies to store your login information if you are registered, and your last visit if you are not. Cookies are small text documents stored on your computer; the cookies set by this forum can only be used on this website and pose no security risk. Cookies on this forum also track the specific topics you have read and when you last read them. Please confirm whether you accept or reject these cookies being set.

A cookie will be stored in your browser regardless of choice to prevent you being asked this question again. You will be able to change your cookie settings at any time using the link in the footer.

Thread Rating:
  • 0 Vote(s) - 0 Average
  • 1
  • 2
  • 3
  • 4
  • 5
TensorFlow, AI, Autopilot, VendeeGlobe
#12
(2023-10-25, 07:33 PM)jim321 Wrote: scrape the web for pics of buoys, boats, ships.
how about ais data too. get course speed "et cetera".
Charts ?

Hi Jim, Thanks for your enthusiasm and the excellent suggestions.

My thoughts... I sail the Solent and I don't think it will take too long to get a good database of pictures of navigational marks. I get tha you could "scrape the web" but as a designer by profession I am aware thats exactly what Midjourney did and I feel that it is wrong to profit even in a non-monetary sense from other peoples data. It could also lead to copywrite issues down the line too.

So I would much rather create the database from my own images and from those sailors interested in the project who are happy to contribute to the collection effort. One such a database is ready it could be used for assisting a self navigating ship to stick to buoyed channels, go the correct side of Cardinal marks etc.
I am thinking about how you would display the information and I guess one way would be to insert data into the NMEA AIS stream. This would also work well for boats that don't have AIS but it might mean a double target for those that did unless you could filter double targets out.

(2023-10-25, 07:51 PM)barrymac Wrote: Nice! 

Here's a product that does the object recognition https://sea.ai/
Starting for the bargain price of 11k euro, plus about another grand for connectivity, up to a glorious 45k for for all bells and whistles. 

They are using NVIDIA® Jetson based platforms 

Perhaps it's possible to crowd source the training data via a signalK plugin? We could really annoy people with a training data captcha asking people to pick the squares with the floating container every log in :-) 

I have a GPU server that could be used for training runs, but I'm quite new to this stuff so I don't really know where to start. 

The coral has object recognition examples available though and the demo videos do look good enough that it would be useful.

Maybe the distance estimate isn't that hard, a camera autofocus type system could potentially give a fairly good estimate.

Hi Barry, Cracking idea! I like you have a GPU server - might need that down the line. I believe you can add one of these Edge accelerators to a Jetson too. I think Raymarine and Sea.ai might be unamused if we managed to get something that works for a couple of hundred dollars. Wink
Reply


Messages In This Thread
RE: TensorFlow, AI, Autopilot, VendeeGlobe - by Hillzzz - 2023-10-26, 10:21 AM

Forum Jump:


Users browsing this thread: 2 Guest(s)