Prev: Aligning Point Clouds
Next: AM
From: ImageAnalyst on 29 Jun 2010 16:44
Like I mentioned, this was one case where you have ambiguity. The
trajectory can still help in cases where the trajectories cross at non-
zero angles, which is probably the majority of cases. Even where the
planes are lining up to take off or land, their trajectories overlap
but they won't be above/below each other in the same frame (time).
You can always count the number of blobs and decide what to do. If
the number doesn't match the prior frame, either some planes have
exited the field of view, entered the field of view, or are
overlapped. You will have to check centroids and trajectories to
determine if you have an "enter/exit field of view" situation, and/or
an "overlap" situation.
From: Dan on 30 Jun 2010 16:53
ImageAnalyst <imageanalyst(a)mailinator.com> wrote in message <ed54cba5-756e-44fc-be3a-593dbb31d466(a)y11g2000yqm.googlegroups.com>...
> Like I mentioned, this was one case where you have ambiguity. The
> trajectory can still help in cases where the trajectories cross at non-
> zero angles, which is probably the majority of cases. Even where the
> planes are lining up to take off or land, their trajectories overlap
> but they won't be above/below each other in the same frame (time).
> You can always count the number of blobs and decide what to do. If
> the number doesn't match the prior frame, either some planes have
> exited the field of view, entered the field of view, or are
> overlapped. You will have to check centroids and trajectories to
> determine if you have an "enter/exit field of view" situation, and/or
> an "overlap" situation.

I think it is going to work to track them frame by frame as you all suggested after all. Thanks for all of your help
First  |  Prev  | 
Pages: 1 2 3 4 5
Prev: Aligning Point Clouds
Next: AM