Dear DAVID Community,

As you are probably aware, the entire DAVID company, team and product range has been integrated into HP Inc.
This forum is now read-only. Relevant content has been migrated to HP and merged into HP's Support Forums on November 1, 2016.

To start new discussions for 3D Scanning, please register and post your new topic at the HP Support Forums

Proper focussing of camera

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

Proper focussing of camera

Postby ArchaeoPhoto » Fri Dec 12, 2008 11:45 am

Hello dear Davidians,

yesterday I started my third testrun of scanning. The result is somewhat ok for handheld scanning and a paper callibration panel which tendend to be in a 90° angle ;-). Well, for now I am satiesfied and the fun is rising with each scan...

During callibration I noticed the following:

I focussed my Webcam to the callibration points (which is a pain with a webcam - I am spoiled with a DSLR). When I got them in focus the callibration was no problem at all.
BUT: When I place my object in front of the callibration panel, it is slightly out of focus. I guess, the aperture of the cam is something around f2.8.
So the laserline is also slightly blurred during scanning, isn't it?!
My question: Does this (dramatically) affect the scan itself?

Next problem is: How do I get a proper texture shot, if the cam is out of focus. If I would touch and move the cam while focussing on the object my callibration would be gone, correct? Then it would not be possible to align scan and texture, am I right?
These are my actual biggest problems, I hope you can give me some suggestions or tipps how to solve it.

Last point: I used grey paper instead of white paper to print my callibration points on. Then I manually set the cam for callibration. The idea is to get a darker background while scanning in order to reduce "reflections" on the callibration panel.
Has anyone experience in using differently coloured backgrounds?

Thank you for reading,

best greetings,

Thorsten
Attachments
Einstein.jpg
That's how he should look like.
einstein_1_tex.jpg
That's the poor quality texture shot. You can see the OOF on Einstein and the "correctly" focussed CPs
test_fusion.JPG
A screenshot from the poisson fusion. Interpolated and Median.
Laser: Green 5mW & 15mW focusable laser; Cam: Color CCD 1280px + Computar 8mm Lens; PC: HP Laptop with 8Gb RAM; Software: David Pro USB, Meshlab, Blender, PS CS4
User avatar
ArchaeoPhoto
 
Posts: 30
Joined: Tue Nov 25, 2008 9:57 am
Location: Bochum, Germany

Postby Bongobat » Fri Dec 12, 2008 12:31 pm

Hi Thorsten,

I always focus on the object being scanned. You will have more depth of field while actually scanning because your aperature will be smaller than when calibrating or taking a texture shot. If it is so shallow that you cannot get a sucessful calibration (printout is blurred) then you need more light and a differnt setting while calibrating.

I hear Gunter saying to turn Einstein on his side to get more resolution!

I used dark construction paper glued to magnetic sheets when I scanned my huruk hai figure. I calibrated with my regular magnetic calibration sheets then removed and placed the dark sheets. I had no problems at all. Using magnetic paint and magnetic sheeting makes differnt setups very easy I think.

hope I helped

Greg
DealExtreme red line laser, Microsoft Lifecam HD-5000
User avatar
Bongobat
Moderator
 
Posts: 1002
Joined: Mon Apr 02, 2007 10:02 am
Location: Ohio, USA

Postby MagWeb » Fri Dec 12, 2008 2:37 pm

puh have some login, logout problems here today....

Turn Einstein on his side to get more resolution!!!!

To solve, or at least ease, your problem of the depth of sharpness, you could try to use a distorted pattern as you can find here:
http://www.david-laserscanner.com/forum/viewtopic.php?t=242 28.Jan.08 and change the calibpoints.dat values according to the pattern.

So the markers are nearer to the cam, in most cases at quite the same distance as the object and you´ll get a better calibration.

As Greg I focus cam and laser on the object too.

Gunter
MagWeb
Moderator
 
Posts: 2499
Joined: Wed Jul 18, 2007 8:48 pm

Postby ArchaeoPhoto » Fri Dec 12, 2008 9:43 pm

Thanks Greg and Gunter for the fast replies.

The focussing problems are now solved. But my Einstein isn't just a "front-scan". The merged image contains of three scans: frontal, 45° left, 45° right.
Do I understand this correctly: in order to getting a high resolution frontal scans should be avoided?

As I wrote, it was just a quick'n'dirty setup. I seriously doubt that my callibration panel had a 90° angle.

@ Gutner: I followed your link. I must admit I didn't yet understand everything of that post. Please be patient, I am an archaeologist and have a natural fear of numbers and anything mathematic ;-)

Soon I'll post some scans of my "ceramic project". In january I'll definetly buy the full version and will present some more high-res scans.
Right now I continiue riding on my learning curve!

Sincerely,

Thorsten
Laser: Green 5mW & 15mW focusable laser; Cam: Color CCD 1280px + Computar 8mm Lens; PC: HP Laptop with 8Gb RAM; Software: David Pro USB, Meshlab, Blender, PS CS4
User avatar
ArchaeoPhoto
 
Posts: 30
Joined: Tue Nov 25, 2008 9:57 am
Location: Bochum, Germany

Postby MagWeb » Fri Dec 12, 2008 10:33 pm

Hi Thortsten

think simple, as a PC: Look at a 3D carthesian coordinate system in 2D: If you want to define points on the planes where one of the three values is equal zero (= the Walls = detection planes). Would you define them near to the origin or as far as you can see, if you want to define the direction of those planes (0Walls = detection planes )? No--- spread your wings and fly away (as Freddy sang) or think turning Einstein´s dimensions.

think I turned to green again --- sorry
Yoda
MagWeb
Moderator
 
Posts: 2499
Joined: Wed Jul 18, 2007 8:48 pm

Re: Proper focussing of camera

Postby Eric Lagman » Sun Mar 08, 2009 3:46 pm

Why does turning the object with the longest part horizontal add more detail? Is it because the camera can see more of where the laser is when it passes? Can someone point me to a thread on this? It would be interesting to see the level of detail with all settings the same, but object turned horizontal. I have been standing most of my objects with the longest part in the y axis. It will be harder to stand things up by laying them on their side, but if it gives a lot more detail it would be worth it.
Eric Lagman
 
Posts: 35
Joined: Tue Oct 21, 2008 1:31 pm
Location: Anderson SC

Re: Proper focussing of camera

Postby Bongobat » Sun Mar 08, 2009 4:20 pm

Hi Eric Lagman,

Suppose your camera's resolution is 640X480 (4:3) and you frame the entire subject maybe a person standing vertical. You only get 480 scanlines of data but if your subject lies down horizontally you can frame them with the wider camera dimension 640 scanlines and hence more resolution. But unless you are not doing planeless scanning you need room for the laser lines on each side and also the object might not fit very easily in the corner so maybe sometimes it is a wash? If you have a camera that is 16:9 you have much more reason to do this.

Hope i am correct,

Greg
DealExtreme red line laser, Microsoft Lifecam HD-5000
User avatar
Bongobat
Moderator
 
Posts: 1002
Joined: Mon Apr 02, 2007 10:02 am
Location: Ohio, USA

Re: Proper focussing of camera

Postby Eric Lagman » Sun Mar 08, 2009 5:10 pm

Yeah I have a quickcam pro 9000 and scan at 960x720. I am using calibration panels so I have to frame those first, and they are taller in the y axis so that is what determines how I frame the scene with my camera. WIth this said then it wont really matter what way I turn the object within the panel space correct? I should get the same resolution regardless shouldn't I? The point of turning sideways is to get the camera as close to the object as possible for higher detail levels correct?
Eric Lagman
 
Posts: 35
Joined: Tue Oct 21, 2008 1:31 pm
Location: Anderson SC

Re: Proper focussing of camera

Postby Bongobat » Sun Mar 08, 2009 5:41 pm

Hi again,

Yes I think you got it now but you dont really have to frame for the calibration printout but it is better for lens distortion removal accuracy if you do. Best problaby to modify the printouts. There have been lots of discussions on modifying them. I myself have not tried it or kept much up in the converstations but I found this: viewtopic.php?f=1&t=242 Since most webcams have a 4:3 aspect ratio shouldnt new David versions come with 4:3 optomized printouts and the software changed to work with those rather than having the user change settings and modify the printouts.

Greg
Last edited by Bongobat on Sun Mar 08, 2009 5:52 pm, edited 1 time in total.
DealExtreme red line laser, Microsoft Lifecam HD-5000
User avatar
Bongobat
Moderator
 
Posts: 1002
Joined: Mon Apr 02, 2007 10:02 am
Location: Ohio, USA

Re: Proper focussing of camera

Postby Eric Lagman » Sun Mar 08, 2009 5:48 pm

Bongobat wrote:Hi again,

Yes I think you got it now. As for the calibration printout there have been lots of discussions on modifying them. I myself have not tried it or kept much up in the converstations but I found this: viewtopic.php?f=1&t=242 Since most webcams have a 4:3 aspect ratio shouldnt new David versions come with 4:3 optomized printouts and the software changed to work with those rather than having the user change settings and modify the printouts.

Greg


I agree. There should be a 4:3 and 16:9 calibration card made available and settings within the software to choose between which one you are using. It makes perfect sense.
Eric Lagman
 
Posts: 35
Joined: Tue Oct 21, 2008 1:31 pm
Location: Anderson SC

Re: Proper focussing of camera

Postby Bongobat » Sun Mar 08, 2009 5:55 pm

You caught me in an edit :oops: Sorry about that.
DealExtreme red line laser, Microsoft Lifecam HD-5000
User avatar
Bongobat
Moderator
 
Posts: 1002
Joined: Mon Apr 02, 2007 10:02 am
Location: Ohio, USA


Return to Camera Calibration

Who is online

Users browsing this forum: No registered users and 1 guest

cron