Prev: Joint Hypothesis
Next: Warning: No display specified. You will not be able to displaygraphics on the screen.
From: ImageAnalyst on 22 May 2010 15:52
You know what? NO ONE is going to finish this project on time. Some
of these images are just too challenging. You'd need a lot higher
level of knowledge than any short class assignment. How is a simple
algorithm going to know the difference between a ship with a bunch of
towers and the cityscape behind it? It won't, unless you build a lot
of real world knowledge into it and that's not easy. It might be
easier if they were videos and moved with respect to the camera and
the rest of the scene. But I seriously seriously doubt anyone in your
class will have every ship in every image automatically outlined any
time soon. Maybe your professor is just trying to see what kinds of
creative ways the students try. You did a passable job with that,
although there could be more things you could demonstrate, and I'm not
sure I know what "conv2 to reduce the number of connected components"
is all about. But I think your professor will give an "A" for effort
and not expect anyone to solve every photo. So just make additional
effort with some more filters and you should get a good grade.
From: Mark Shore on 22 May 2010 16:23
"Mark Shore" <mshore(a)magmageosciences.ca> wrote in message <ht6iot$k1g$1(a)fred.mathworks.com>...
> "Emre " <jasonborn(a)windowslive.com> wrote in message <ht6i65$bko$1(a)fred.mathworks.com>...
> > Hi,
> > I'm working on a project to detect ships at horizon.
> > But, first I have to detect horizon and then work on object above the horizon.
> > I'd be glad to receive any help.
>
> Google it, or work it out from scratch. Only simple trigonometry is required and it will be trivial to implement in MATLAB.

Never mind this comment. From your first post I assumed you were referring to the classic problem calculating the distance to an object of height A from a viewpoint of height B, with the line of sight tangent to the Earth's surface. As far as your much different project goes, ImageAnalyst has said more than I could.
From: ImageAnalyst on 22 May 2010 17:14
On May 22, 4:23 pm, "Mark Shore" <msh...(a)magmageosciences.ca> wrote:
> Never mind this comment. From your first post I assumed .....

------------------------------------------
Just goes to show you how much time and effort would be saved if
people would only post their images FIRST. It happens all the time so
I shouldn't be surprised, but I always am at people who ask how to
process and image without showing people the image.
First  |  Prev  | 
Pages: 1 2
Prev: Joint Hypothesis
Next: Warning: No display specified. You will not be able to displaygraphics on the screen.