home *** CD-ROM | disk | FTP | other *** search
- Path: sparky!uunet!munnari.oz.au!manuel.anu.edu.au!dubhe.anu.edu.au!barnard!andrewl
- From: andrewl@cs.anu.edu.au (Andrew Longhorn (Vac Scholar))
- Newsgroups: comp.robotics
- Subject: high resolution 3D imaging / robotics vision?
- Date: 16 Dec 1992 22:24:27 GMT
- Organization: Computer Science Department, ANU, Australia
- Lines: 61
- Distribution: world
- Message-ID: <1goaarINNkqc@dubhe.anu.edu.au>
- Reply-To: andrewl@cs.anu.edu.au
- NNTP-Posting-Host: barnard.anu.edu.au
-
-
- I thought I would ask this, Does anyone know if much progress/work/theorising
- has been done on 3D depth mapping?
-
- What I am trying to get a phrase for is some sort of system that you point
- at something and it returns a 2d map of z-values corresponding to the distance
- from the sensor of the first thing it hits. (sort of like a 2d version of
- those fish sensor things they use in fishing), y'know, they point it downwards
- beneath the boat and get something like:
-
- . x
- . x
- .. . .
- . .......... ..... ...... . <- sea bed
- .
-
- where: x - is possible fish schools.
-
- Anyway, what I need is some system for determining the distance to objects
- within about a 100 degree cone (or sq. base pyramid) eminating from the
- sensor location:
-
- X <- sensor
- / \
- / \
- / \
- / \ <- sensor range and area
- / ....... \
- /.. ..\
- .. ..
- ...........
-
- Hopefully returning a 2d grid with the distance information and some special
- number (or really large) for out-of-range values. so if you point
- it down at a table with a block on it you may get something like:
-
- 11 10 10 10 9 10 10 10 10 11
- 11 10 10 10 9 6 10 10 10 11
- 11 10 10 6 5 6 6 10 10 11
- 11 10 6 6 5 6 6 10 10 11
- 11 10 10 6 5 6 6 6 10 11
- 11 10 10 6 5 6 6 10 10 11
- 11 10 10 10 5 10 10 10 10 11
- 11 10 10 10 9 10 10 10 10 11
- 11 10 10 10 9 10 10 10 10 11
-
- ^ ^ ^ -- these would probably be results of the
- point-radiating nature of the sensor - helpful
- for my intended application.
-
- Anyway, you all get what I mean. I'd imagine this sort of thing would be
- really helpful in robot/computer vision systems, and other areas.
-
- ANY leads on technology like this, ranging from heresay/legend to prices and
- distributors, would be greatly appreciated.
-
- Regards,
-
- Andrew
-
- andrewl@cs.anu.edu.au
-