Subject:
|
Re: Robotic Vision (was: Video image analisys)
|
Newsgroups:
|
lugnet.robotics
|
Date:
|
Fri, 4 Feb 2000 19:16:37 GMT
|
Original-From:
|
Joel Katz <jmk@acut*spamcake*.com>
|
Viewed:
|
2171 times
|
| |
|
|
James:
You wrote:
> I have seen this program (NIH) at a workshop, and it was pretty incredible.
> I checked on the PC version (Scion makes this) and it is very similar. It
> says that it can measure (distances?) Could this be used as "vision" for a
> robot? Could this be used to "tell" a robot where it is?
>
> Lots of questions, and fairly excited about the future!
Absolutely, something like NIH can be used either for determining
where your robot is in the world (eg. with a fixed camera looking down at
your robot), or telling where things are relative to your robot with a
robot mounted camera.
There are a number of important issues to address, like
1) calibrating from Image Space which is usually in pixels, to world space,
which has real world units, or robot space, which is a coordinate system
attached to your robot.
2) figuring out what features can be measured reliably to give you useful
measurements in a non-structured scene
3) addressing the issue of getting 3 dimensional measurements from an
inherently 2 dimensional detector.
There are many solutions to each of these problems and you can have
a lot of fun exploring the possibilities.
--joel
Joel Katz, Principal Engineer --- (603)-577-5895
Acuity Imaging, 9 Townsend West, Nashua, NH 03063 --- (603)-577-5965 (FAX)
visit our Web Site at >>> www.acuityimaging.com <<<
|
|
Message is in Reply To:
 | | Re: Robotic Vision (was: Video image analisys)
|
| I have seen this program (NIH) at a workshop, and it was pretty incredible. I checked on the PC version (Scion makes this) and it is very similar. It says that it can measure (distances?) Could this be used as "vision" for a robot? Could this be (...) (24 years ago, 4-Feb-00, to lugnet.robotics)
|
2 Messages in This Thread:
 
- Entire Thread on One Page:
- Nested:
All | Brief | Compact | Dots
Linear:
All | Brief | Compact
|
|
|
|
