Alignment problems

Write here about construction, ideas, equipment, tips n tricks etc. related to structured light scanning
3dxcan
Posts: 67
Joined: 02 Jun 2017, 17:21

Re: Alignment problems

Post by 3dxcan » 24 Aug 2017, 19:58

im getting elipses away from the center, instead of circular target points on the calibration panel. its on the side thats not facing the camera directly. the other side facing the camera looks ok.
Image

What do you suggest? to eliminate this?

Curiousjeff
Posts: 121
Joined: 16 Nov 2016, 22:31

Re: Alignment problems

Post by Curiousjeff » 24 Aug 2017, 20:55

It's hard to judge from a screenshot with that side of panel on an angle.

Questions:

What are the angles between camera and projector (horizontal and vertical)
Are you scanning only in horizontal or both ?
What version of David ?
What camera ?
How is your panel made ?

Jeff

3dxcan
Posts: 67
Joined: 02 Jun 2017, 17:21

Re: Alignment problems

Post by 3dxcan » 25 Aug 2017, 18:11

Curiousjeff wrote:
24 Aug 2017, 20:55
It's hard to judge from a screenshot with that side of panel on an angle.

Questions:

What are the angles between camera and projector (horizontal and vertical)
Are you scanning only in horizontal or both ?
What version of David ?
What camera ?
How is your panel made ?

Jeff
23UX236, LG PF1500W, my setup is horizontal, settings is "both", David 5.2, Panel made by myself ( good enough as drndadoo@HP says) , One thing I've noticed is that when the camera captures distance away from the main focus point, those farway areas are badly distorted. (See the winodw scan, the yellow scan is badly distorted, it captured the near side of the window accurately, and aligned well, seems like i forgot to hit the inverse selection button before i hit delete and the good part of the yellow scan is gone in the image). so one part of the problem is depth of field. wouldn't a better lens help? (current lens computar 1614mp2), isnt this due to Camera shutter being rolling type?

Image
Attachments
blueshot-outof-focus.PNG

NARMELLA
Posts: 13
Joined: 19 Apr 2017, 05:53

Re: Alignment problems

Post by NARMELLA » 26 Aug 2017, 06:28

Could you please share those scans?
Wanna to check .

3dxcan
Posts: 67
Joined: 02 Jun 2017, 17:21

Re: Alignment problems

Post by 3dxcan » 11 Sep 2017, 02:27

Update to my status, i tried with a global shutter camera and i got much better results, however i think if i can get david to project the patterns slower, i can get decent results with rolling shutter too. I have to try by increasing exposure time i think.

Is there any other ways to make david project the patterns slower?

Out of focused areas of larger objects is also problematic as tge scan of those areas have muxh lower quality. Thus if theres an automatic way of removing those, or make the the depth of field larger, that would also help.

I know stopping down tge lens results in larger depth of field but the effect isnt enough and tge picturer loses sharpnesss,

Does a better lens like the schnider help to get a larger depth of field?

User avatar
Micr0
Posts: 184
Joined: 15 Nov 2016, 15:20

Re: Alignment problems

Post by Micr0 » 11 Sep 2017, 02:52

One of the problems that I have found in stopping down a lens to try to get a larger depth of field is that that introduces refractive error that David doesn't seem to compensate for. As per our discussion off line, I did some tests today with a setup calibrated to 200mm panels. Your assertion seems to be correct in that the the depth of the scan envelop seems to be very shallow. Shallower than I had previously thought. The scan depth seemed to be only about ~150mm (+- 75mm and maybe even less) before the accuracy became questionable.
µ

3dxcan
Posts: 67
Joined: 02 Jun 2017, 17:21

Re: Alignment problems

Post by 3dxcan » 13 Sep 2017, 07:14

Micr0 wrote:
11 Sep 2017, 02:52
One of the problems that I have found in stopping down a lens to try to get a larger depth of field is that that introduces refractive error that David doesn't seem to compensate for. As per our discussion off line, I did some tests today with a setup calibrated to 200mm panels. Your assertion seems to be correct in that the the depth of the scan envelop seems to be very shallow. Shallower than I had previously thought. The scan depth seemed to be only about ~150mm (+- 75mm and maybe even less) before the accuracy became questionable.
yes thats what I was talking about. Narmela can achieve accurate scans because he's using DSLR and Lets scan, Is there a way to achieve the same level of depth of field using industrial camera or its the limitation of sensor/lens size?

SilverGhost
Posts: 9
Joined: 07 Mar 2017, 07:47

Re: Alignment problems

Post by SilverGhost » 13 Sep 2017, 12:27

3dxcan wrote:
13 Sep 2017, 07:14
yes thats what I was talking about. Narmela can achieve accurate scans because he's using DSLR and Lets scan, Is there a way to achieve the same level of depth of field using industrial camera or its the limitation of sensor/lens size?
The deth of field for the DSLR will always be smaller because of it's sensor size: the bigger the sensor - the smaller the depth of field.

3dxcan
Posts: 67
Joined: 02 Jun 2017, 17:21

Re: Alignment problems

Post by 3dxcan » 26 Sep 2017, 19:51

Any new reslts or development toward minimizing distortions for larger parts?

spark
Posts: 40
Joined: 06 Mar 2017, 20:44

Re: Alignment problems

Post by spark » 27 Sep 2017, 07:06

i am really exited about the calibration routine of flexscan3d.
The DOF is also beeing considered...

https://www.youtube.com/watch?v=ovnn6sSlSHk

Post Reply

Who is online

Users browsing this forum: No registered users and 1 guest