Fiducial marker code
- From: "David Olsson" <vonolsson@xxxxxxxxx>
- Date: 23 Oct 2006 22:16:54 -0700
I am working on a system which will allow me to detect square fiducial
markers containing visual codes represented by binary (black or white)
squares evenly distributed on a grid on the marker in photographs
and/or video frames. Currently, I successfully identify candidate
markers and sample their area for the marker codes. I however have not
been to come up with any successful coding scheme. The problem I have
had with coding schemes I have tested is that codes may be confused
with other codes if they are rotated or mirrored or that the frequency
of false positives gets very high.
Currently I am using a 6x6 grid and thus have 36 bits to work with. I
have been trying ot come up with schemes using 6-bit CRC checksum to
avoid false positives and rotation/mirror invariant codes but with no
luck.
I would be very grateful if someone could point me in the right
direction!
Best regards
David Olsson
.
- Follow-Ups:
- Re: Fiducial marker code
- From: serg271
- Re: Fiducial marker code
- From: Rival
- Re: Fiducial marker code
- Prev by Date: Re: Gabor transform and Gabor filter
- Next by Date: Re: Fiducial marker code
- Previous by thread: BUG in Java ImageIO,problem in JPEGImageDecoder,can't read or write image
- Next by thread: Re: Fiducial marker code
- Index(es):
Relevant Pages
|
