PDA

View Full Version : Summary of replays: scanning laser source/video projector



tpribanic57
04-23-2003, 07:59 PM
Dear All,

here is a summary of replies to my question about laser source and video-projector calibration. The last contribution is from different discussion group but it is very descriptive in practical sense and it is included here also. I thank you all.

Regards, Tomislav.

Tomislav Pribanic, M.Sc., EE
Department for Electronic Systems and Information Processing
Faculty of Electrical Engineering and Computing
3 Unska, 10000 Zagreb, Croatia
tel. ..385 1 612 98 67, fax. ..385 1 612 96 52
E-mail : tomislav.pribanic@fer.hr


ORIGINAL posting:
the general idea of projecting laser line and/or point on a subject in space, capturing it with cameras and finally performing 3D reconstruction is well known. Although the laser source is not the only alternative I was wondering if any of you can recommend me a product on market which can scan throughput the space a laser line and/or point. Some sort of scanning control would be welcome as well, but (perhaps) not mandatory.
My second question is kind of related to the first one: has anyone
tried to calibrate video-projector, namely to find its calibration
parameters in space in similar fashion as with 'normal' camera?

REPLAYS:

Dear Tomislav,

If you are looking at small objects that can be clamped onto the measuring
device, I can refer you to my MSc thesis (Couillard, S. University of
Calgary (Canada), Department of Mechanical Engineering, 2002). I looked at
the deformation of articular cartilage by measuring the topography of the
surface before and after a load was applied. I used an off-the-shelf laser
scanner (MicroScan Laser Profilometer, LMI Technologies, Southfield, MI,
USA).

You can also find conference abstracts that relate to this laser scanner in:

Couillard, S., Herzog, W. 2002 Cartilage Deformation of the Feline
Patellofemoral Joint Obtained from Laser Scanning. Proceedings of the IVth
World Congress of Biomechanics

Couillard, S., Herzog, W. 2001 Accurate Articular Surface Measurement Using
Laser Scanning. Proceedings of the XVIIIth Congress of the International
Society of Biomechanics p. 292-293.

Couillard, S., Herzog, W., Péry, V. 2000 In Situ Cartilage Deformation
Resulting From Physiological Loading of the Feline Patello-Femoral Joint.
Archives of Physics & Biochemistry 108(1-2): 179.


Good luck!

Sylvain
*******************

Hi Tomislav
Depending on what you need there are multiple technologies available.
The other parameter will be scanning volume (distance from source * size
of object). So we are using a Minolta 3D Vivid 900 non-contact
photographic laser scanner. And I am quite happy with its speed and
resolution. We can go down to 250 micron for slice thickness and
150micron for inline pixel size. If you go this site
(http://www.rapidform.com/partner/part_main.htm) there are whole bunch
of scanner info and links...
It will depend on your application but I wouldnt recommend point
scanners...
Best,
Murat
*******************

Dear Tomislav,

we did, there is no problem indeed if you project a grid with the projector
on a plane moved in the working volume (Zhang approach)

Alberto Borghese
***************************

----- Original Message -----
From: Frank Wu
To: OpenCV@yahoogroups.com
Sent: Wednesday, April 16, 2003 6:34 PM
Subject: Re: [OpenCV] video projector calibration


Hi, Tomislav:

That's very interesting function. Here is my opoinion:

1. First, calibarte the camera
2. Image a flat panel with chess board and find its space postion by cvFindExtrinsic...... Function.
3. Put a white paper to cover the chessboard, but still keep the board on its old position.
4. Project a chessboard iamge from video-projector into the white paper, find the corners, then you know the 3D position relative the flat panel (the Z should be always zero). Build the corrsonding of 3D points in the flat boards and 2D points in the chessboard image.
5. Repeat 2-4 several times, use the cvCalibrateCamera to find the intrinsic of the video-projector.

Frank Wu

---------------------------------------------------------------
To unsubscribe send SIGNOFF BIOMCH-L to LISTSERV@nic.surfnet.nl
For information and archives: http://isb.ri.ccf.org/biomch-l
---------------------------------------------------------------