I strongly quote in_vino_vomitus (what the heck of a nickname, btw :D)
I really like your commenting stye in your videos (found some funny stuff in your STAKER let's play too, lol), please don't change :D

Not a problem :haha: in_vino_vomitus' context clues made it sound like you were contrasting your appreciation of my stuff with that quote. It's fine, though. If you like it, good. That change was gonna be a long and hard one (that's what she said).

Also, I'm glad you checked out S.T.A.L.K.E.R., banryu :D My other videos don't get as much love haha

If you haven't seen, I just started the FFG tuts so I will get to that shortly :D

Nice one :)

What I do is stick a tracker on all the contacts and go into the navigation map and mark all the current bearings. Then turn to eliminate any ghost contacts.

Just a test on my tool. Used the situation in the video at around 13 mins.

http://www.subsim.com/radioroom/pict...ictureid=10302

Cool, I saw your thread about the tool. What are the results of the test?

Works well most of the time, but in some cases (such as the above one) when there're so many congested intersections between lines, the opencv line detector could fail to detect some portion of the line. But still it can give you some reliable TMA solutions.

And I couldn't find a way to add speed constraints since it's using image pixel coordinates. All I can do is to draw the solution on the image so you know where it is.

In the above example, the tool has found two solutions, displayed in green. And it seems the longer one is the 10-knot target we want.