You might also find this site interesting:

http://www.muellerr.ch/engineering/laserscanner/default.htm

as his setup works without a projector (which I thought was a tad 
expensive).

About a year ago I made my own prototype...

As I didn't have a stepper motor I just span the the laser stripe by hand.

I wrote some quick code to work out the angle the laser is pointing from 
the image.

I got my laser stripe from a cheap Jigsaw - Yes I was walking around a 
hardware store and thought .oO "cool there's a cheap laser stripe with a 
free powertool attached" :-)

As a prototype it seemed to work but I only saved the range data as an 
image that got rather corrupted by the compression.

I've since got fedup with Java on Windows and this week have set myself 
the task of rewriting in C on Linux.

Here's a scaled down version of the range image I got.
I tried yesturday to create a height field out of it but it didn't 
really go to plan :-(

"Knightstar" <kni### [at] yahoocom> wrote:
(in a year old thread which sort of throws my mail reader :-) )

 > Hi all update here is some info on some software and web sites to look
 > At. At this time I am working on the scaner and will have some step by
 > step photo's.

 >       PUBLIC DOMAIN 3-DIMENSIONAL DATA ACQUISITION DESIGN AND SOFTWARE

 >        http://www.engr.uky.edu/~lgh/3dDA/hb2003.htm

 >      3D Scanners, Digitizers, and Software for making 3D Models

 >        http://www.netheaven.com/~simple3d/

 >                                 Happy scaning,
 >                                                 knightStar

Post a reply to this message

Attachments:
Download 'rangeimage.png' (52 KB) Download 'heightfield.jpg' (35 KB)

Preview of image 'rangeimage.png'

Preview of image 'heightfield.jpg'

Heheh...what a coincidence. I started making a 3d scanner not too long ago
for use in pov.
It's nearly complete, although I've decided to move the laser beam with the
camera along a linear path rather than sweeping it as in the link (that's
where I'm at now, building the positioner assembly). The first method I
used was to rotate the object on a turntable to make a cylindrical height
field, but now I want to scan the faces of large objects so I can get
accurate models (such as a head from an automobile engine) into Solidworks.
Still having HUGE problems with the Canon camera SDK. It's so unstable. I
think I'll have to limit the camera type to Logitech webcams. Too bad too.
Could have been many megapixels per image taken for each laser line, but,
looks like it will be 640x480 if I can't get it working. Resulting height
field will have much less resolution/accuracy :(.

Post a reply to this message