View Full Version : video projector vs. laser

04-23-2003, 08:18 PM
Dear All,

thank you for replaying to my previous question about video-projector calibration Here is my additional concern when using video-projector to project a strip of light on a subject for the purpose of 3D shape reconstruction (i.e. 3D scanners). Such a systems, among other things, request finding the edge of the projected strip. From the brief overview of such systems on the Web seems to me they all project strip of light on the surface which has more or less same color. In that case (considering the known definition of edge; 'place where abrupt change of gray level intensity occurs') finding the strip edge is relatively easy. However lets consider the colorful surface (org.jpg). Projecting the strip of light in a dimmed ambient light (strgrb,jpg) and converting the image into graylevel one (strgray.jpg) shows that transition from light to dark is not the same along the edge of the stripe (i.e. looking at different rows of image from top to bottom). Consequently finding the edge pixels and/or setting up some threshold value for the edge is harder. Moreover, some areas within the stripe may occur only slightly lighter than areas outside of the projected stripe. Under such a circumstances how can one find reliable algorithm to extract edge of projected stripes? Are 3D scanners with video projector inherently restricted for surfaces with more or less have same (shades) colors? Or should the video projector light source be intensive enough to overcome those difficulties? Would system which would use laser source strip be more efficient since laser light is monochromatic and as a rule more intensive?
When trying to reconstruct the shape of the face these problems might occur with eyelashes,eyebrow and hair (colgray.jpg) .

Regards, Tomislav.

P.S. Images itself are not included in e-mail since this list is setup to reject them. I would be happy to send those directly to anyone who is interested.

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

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