Dear DAVID Community,

In 2016, DAVID and this forum have been integrated into HP Inc.
This forum has been read-only since then, and will be switched off by the end of 2019.

To everyone who has contributed here, we thank you for all your work and for the constructive, helpful and friendly atmosphere you have created!
Please continue to discuss on the HP Support Forums.

David camera calibration vs Open CV

The place for questions, problems, comments and tips regarding the camera calibration.

David camera calibration vs Open CV

Postby pognibene » Tue Mar 06, 2012 8:53 pm

Hi all,
It looks like David camera calibration is based on the Tsai model, or a derivative of it.
Values produces in the calibration XML file look very similar to Tsai parameters indeed.
I'm pretty sure somebody around played with Open CV, that integrates some support for camera calibration as well (but not based on a 90° corner, rather on a flat checkerboard).
Are Tsai intrinsic coefficients produced by David the same than what I would get with Open CV? Is the Open CV calibration more accurate than David's one, or the opposite?

Ok, that's a bit off topic and maybe an internal implementation detail, but still an answer would be interesting.

Thank you all for your help.

Pascal
pognibene
 
Posts: 12
Joined: Sun Nov 13, 2011 7:15 pm

Re: David camera calibration vs Open CV

Postby manish » Wed Mar 07, 2012 4:02 pm

hi,

u asked my question?
i was also going thru article for
calibration of projector & camera by matlab toolbox.
found interesting..... question raised........
which calibration better David or open CV

Regards
Manish
manish
 
Posts: 120
Joined: Sat Oct 30, 2010 6:18 pm
Location: mumbai

Re: David camera calibration vs Open CV

Postby Bolte83 » Mon Nov 05, 2012 11:42 pm

HI

I made a test in our Photogrammetrie Labor at University ( self developed Calibration Software). 3D Calibrationpattern - take a lot of Photos (more then 24) from different positions (dont change focus -> same distance) rotate the camera und change Viewangle. For more details feel free to ask.

InnerOrientation
DAVID Monochrom Cam. 1024x768 Photogrammetrie Software
Code: Select all
                     cx | xh        cy | yh       f | c       sx | sxy           Verzeichnungsparameter der Kailbrierung         
David Cam 1024x768   5,11490441   -11,1129266   1270,095600   0,999369318   9,7049E-08                                A1 Photosoftware
David Cam 1024x768   5,35168237   -11,4922754   1271,146300   0,999345909   1,1685E-07   7,3289E-14                   A1 A2 Photosoftware
David Cam 1024x768   5,55422248   -15,0121536   1269,724750   0,999393818   9,8776E-08   1,8320E-07   -1,0327E-06     A1 B1 B2 Photosoftware
David Cam 1024x768   4,94125473   -11,3576470   1262,393660   0,997072127   12 Stück Angabe siehe Textdokument        Chebichew Photosoftware
David Cam 1024x768   6,64410086   -11,2458285   1316,609370   0,998359656   Ohne         Ohne                         Photosoftware
David Cam 1024x768   5,16151292   -11,6858774   1272,496250   0,999257922   9,6802E-08                                A1 Photosoftware
David Cam 1024x768   524,275164   366,2347928   1293,829106   0,995645587   9,1652E-08                                David Software

Notice that the Origin of Coordinate is different ! DAVID = Uper Left Corner // Photo-Software = Center of Image

Easly you can see that the center of a optical distortion differ up to 6-8 Pixel in X and Y direction and has a big impact. The difference in the Focal length has no significant influence on the measurement.
All calibarations are calculatet 5 Times to see if they are stable.
Student: Geodäsie & Geoinformation / DAVID-CAM-1-M / Red & Green FocusLaser
Bolte83
 
Posts: 7
Joined: Mon Jul 09, 2012 7:55 am
Location: Bonn


Return to Camera Calibration

Who is online

Users browsing this forum: No registered users and 0 guests

cron