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

The Great Lines in Scans Experiment.

Which camera works? Where can you buy an adequate laser? And more...

The Great Lines in Scans Experiment.

Postby micr0 » Fri Apr 17, 2015 3:32 pm

Many including my self have complained about parallel lines in our scans. After no comment from David tech people we have been left to guess at what causes them. I hypothisized as have others that is is the fan in the projectors that is inducing vibration into the light patterns and thus causing these problems in the scans. So in the interest of smoother surfaces scans I've been wanting to experiment with a projector and see if I could eliminate fan (or other induced) vibration. I started my David project with an Acer K330 as my projector. That rig had a very pronounced scan line problem and when I switched to an Optoma ml550 projector it was a great improvement. An improvement but the lines were still there. Since I'm not using the K330 anymore I decided to take it apart and see if I took the fans out if that solved the line problem. So I took it apart. First off, the K330 Has 3 fans. one that brings air in and blows it over the power supply board, the next is a twin fan that moves air over 2 heat pipes that are attached to the light sources. the twin van vibrates horribly when energized and is very poorly covered for vibration damping. also it and the light source are ridgidly mounted to the very thin plastic projector housing and it is clear the any vibration from the fan is easily transmitted through the housing to the emitter assembly. I tried to turn the projector on the the fans disconnected but the control immediately throws a temp error. since the lights aren't even on and there is no heat being generated the system must be sensing the lack of current draw on the fan circuits and throwing the error. Similarly when the LEDs are on they heat up very quickly so in order to keep from damaging them I will have the fans directly on the heat pipes. What I plan to do is make a heavy metal base to mount the light source and lens assembly on thus increasing it's effective mass. Then I plan to mount the fans via a rubber band like suspension so that they blow over the cooling fins of the heat pipe but are further isolated from the emitter assembly by having the base of that suspension mounted to a separate tripod to the floor. I'll let you know how it works out as soon as I get a bit of time to make the parts. Stay tuned.


µ
Attachments
P1110386.JPG
µ
micr0
 
Posts: 315
Joined: Wed Mar 19, 2014 3:54 pm
Location: NYC

Re: The Great Lines in Scans Experiment.

Postby micr0 » Fri Apr 17, 2015 3:46 pm

A description of the inside of the Acer K330:
Attachments
K330.jpg
µ
micr0
 
Posts: 315
Joined: Wed Mar 19, 2014 3:54 pm
Location: NYC

Re: The Great Lines in Scans Experiment.

Postby Curiousjeff » Sun Apr 19, 2015 12:52 pm

I share your frustration regarding the cause of the waves/strips in the scans.

I also want to try different hardware adjustments, but I thought it was a good idea to study the images that David is capturing and try to identify the problem before.

I setup the folder: SL-TestSequence

TEST 1: All images from a single vertical scan (27 images) of a newspaper without the light of the projector touching the paper.

1) After calibration, I scanned a flat page with printed text. The projector is prevented from lighting the paper (but it is ON, with fan turning). In this case, all images should be the same without any projector illumination. I then loaded all the images into Photoshop, placed them one on top of another to see if there was any shaking / misalignment but this did not seem to be the case. There were differences, but they seem to be frequency/shutter related. Some images were brighter then others.

TEST 2: All images from a single vertical scan (27 images) of a newspaper with the same pattern projected on all 27 images.

2) I scanned a flat page with printed text. The projector is ON but it is projecting a single pattern (in David, I switch the projector screen to 1 and loaded a pattern as my Windows desktop background). In this case, all scan images will be photos of the same paper with the same illumination. I then loaded all the images into Photoshop, placed them one on top of another to see if there was any shaking / misalignment but this did not seem to be the case. Same as previous test.

But the actual scan of a 3D items does have strips ?

I am not sure now that the fan is the culprit ?

I did notice something else unusual. I scanned twice the same item with the same settings and miracle..the second scan was much better.

I am puzzled.

If the cause was some vibrations, I imagine that it would not be too difficult for David to realign the images before the "interpretation process", for example by placing in all the patterns a crosshair and placing a physical marker in the scene at the same place as the projection crosshair. This would allow correction of camera and/or projector shaking.

If the cause is shutter/frequency, I imagine some algorithm should be able to deal with it prior to processing (?)

Last solution, a specific algorithm in David only to deal with strips but without softening the rest of the texture.

I would love to see a series of photos that give a nearly perfect scan.

Jeff
Panasonic GH4 4K with lens 12-35, Inogeni 4K to USB3 hdmi capture device, projectors: Acer K135/Acer K11/NEC V260X, Canon 60D, macro 60 mm, MagicLantern
Curiousjeff
 
Posts: 230
Joined: Fri Feb 27, 2015 2:17 pm
Location: Geneva, Swizerland

Re: The Great Lines in Scans Experiment.

Postby Curiousjeff » Sun Apr 19, 2015 1:06 pm

Pattern projected

+

Same scan twice
Attachments
image no 008 type 2 freq 008.jpg
Projected pattern
2015-04-19_14-02-41.jpg
Panasonic GH4 4K with lens 12-35, Inogeni 4K to USB3 hdmi capture device, projectors: Acer K135/Acer K11/NEC V260X, Canon 60D, macro 60 mm, MagicLantern
Curiousjeff
 
Posts: 230
Joined: Fri Feb 27, 2015 2:17 pm
Location: Geneva, Swizerland

Re: The Great Lines in Scans Experiment.

Postby Curiousjeff » Sun Apr 19, 2015 5:34 pm

Forgot to mention my equipment

Canon 60D
Acer K135
Panasonic GH4 4K with lens 12-35, Inogeni 4K to USB3 hdmi capture device, projectors: Acer K135/Acer K11/NEC V260X, Canon 60D, macro 60 mm, MagicLantern
Curiousjeff
 
Posts: 230
Joined: Fri Feb 27, 2015 2:17 pm
Location: Geneva, Swizerland

Re: The Great Lines in Scans Experiment.

Postby Curiousjeff » Sun Apr 19, 2015 10:50 pm

Some improvement.

All default settings in David except

Phase shift -> Multipler : 4
Noise reduction 2
Minimum contrast 2

Vertical setup, around 17 degré angle (camera in front, projector on top)

Fine adjustment of the shutter speed using MagicLantern on the Canon 60D

(Movie mode, ISO 200, Fstop 14, Shutter 1/49 (279 degree)

Calibration 109 mm
Attachments
2015-04-19_23-42-01.png
Panasonic GH4 4K with lens 12-35, Inogeni 4K to USB3 hdmi capture device, projectors: Acer K135/Acer K11/NEC V260X, Canon 60D, macro 60 mm, MagicLantern
Curiousjeff
 
Posts: 230
Joined: Fri Feb 27, 2015 2:17 pm
Location: Geneva, Swizerland

Re: The Great Lines in Scans Experiment.

Postby Curiousjeff » Sun Apr 19, 2015 10:53 pm

It is hard to determine if the fan contributes to the wave / strip problem, but there is no doubt that the frequency difference between the projector and the camera is the major cause.
Panasonic GH4 4K with lens 12-35, Inogeni 4K to USB3 hdmi capture device, projectors: Acer K135/Acer K11/NEC V260X, Canon 60D, macro 60 mm, MagicLantern
Curiousjeff
 
Posts: 230
Joined: Fri Feb 27, 2015 2:17 pm
Location: Geneva, Swizerland

Re: The Great Lines in Scans Experiment.

Postby micr0 » Mon Apr 20, 2015 2:01 pm

Curiousjeff wrote:Some improvement.

All default settings in David except

Phase shift -> Multipler : 4
Noise reduction 2
Minimum contrast 2

Vertical setup, around 17 degré angle (camera in front, projector on top)

Fine adjustment of the shutter speed using MagicLantern on the Canon 60D

(Movie mode, ISO 200, Fstop 14, Shutter 1/49 (279 degree)

Calibration 109 mm


Can you post a .jpg of what your scanning so we can get an idea of the surface texture of the object in the scans?


µ
µ
micr0
 
Posts: 315
Joined: Wed Mar 19, 2014 3:54 pm
Location: NYC

Re: The Great Lines in Scans Experiment.

Postby Curiousjeff » Mon Apr 20, 2015 4:53 pm

It's a knife holder from the brand Bodum.

The surface is mostly smoth plastic.

https://www.google.ch/search?q=bodum+kn ... 200%3B1200
Attachments
Bodum.jpg
Panasonic GH4 4K with lens 12-35, Inogeni 4K to USB3 hdmi capture device, projectors: Acer K135/Acer K11/NEC V260X, Canon 60D, macro 60 mm, MagicLantern
Curiousjeff
 
Posts: 230
Joined: Fri Feb 27, 2015 2:17 pm
Location: Geneva, Swizerland

Re: The Great Lines in Scans Experiment.

Postby micr0 » Mon Apr 20, 2015 5:47 pm

I got a chance to test out my theory. I was able to calibrate the scanner with the K330 and scan my 190mm calibration panels which are white glass and very smooth. I remounted the emitter and rigged the system so that I could remove all 3 fans by both suspending them separately and shutting them completely off during the scan. It made no difference. The lines in the scans are just as pronounced with the fans on as they are with them off. Back to square one. As to why the Optoma would produce better scans than the Acer is now a mystery. perhaps it has to do with the frame sync of the camera vs projector as mentioned. The way the DLP refreshes? I'm at least pretty sure that it's not fan induced vibration.

Suggestions?

It seems that most 3D ready DLP projector (of which the Optoma is one) send a sync signal via "DLP Link" for synchronizing active shutter 3D glasses. I wonder if there is a way to use this to trigger the cameras that have a trigger in capability? One may not want to sync every frame. And with the cameras that are running a frame rates that are not similar to the projectors there is obviously an issue but it bears thinking about......
µ
micr0
 
Posts: 315
Joined: Wed Mar 19, 2014 3:54 pm
Location: NYC

Re: The Great Lines in Scans Experiment.

Postby Curiousjeff » Mon Apr 20, 2015 7:38 pm

I am sorry for the long first email, but all those tests with Photoshop made me believe that the problem was not vibrations related. I preferred this approach before I took apart my brand new projector.

Since I am using a Canon DSLR (60D) as camera, I am playing around MagicLantern. It gives tremendous control on shutter speed and frame rate. I can see on the camera screen the improvements before I even scan.

Depending on the adjustment, the frequency lines nearly disappear.

It's not perfect, but getting better.

You suggestion regarding a real sync sounds great.

Its frustrating, because its obvious that near perfection is possible, but the David team would need to pitch in.
Attachments
2015-04-20_20-27-22.png
Panasonic GH4 4K with lens 12-35, Inogeni 4K to USB3 hdmi capture device, projectors: Acer K135/Acer K11/NEC V260X, Canon 60D, macro 60 mm, MagicLantern
Curiousjeff
 
Posts: 230
Joined: Fri Feb 27, 2015 2:17 pm
Location: Geneva, Swizerland

Re: The Great Lines in Scans Experiment.

Postby Curiousjeff » Mon Apr 20, 2015 8:10 pm

Ok, now I feel stupid.

I had changed the projector resolution. I was not using the native resolution.

Here is now my best result

Camera settings:

Frame rate exactly 30 frame per second
Shutter speed 24
Fstop 7
ISO 100
Attachments
2015-04-20_21-07-10.png
Panasonic GH4 4K with lens 12-35, Inogeni 4K to USB3 hdmi capture device, projectors: Acer K135/Acer K11/NEC V260X, Canon 60D, macro 60 mm, MagicLantern
Curiousjeff
 
Posts: 230
Joined: Fri Feb 27, 2015 2:17 pm
Location: Geneva, Swizerland

Re: The Great Lines in Scans Experiment.

Postby Curiousjeff » Mon Apr 20, 2015 8:12 pm

No more waves :)

I am in Europe.

Projector is at 1280 x 800 @ 60Hz
Panasonic GH4 4K with lens 12-35, Inogeni 4K to USB3 hdmi capture device, projectors: Acer K135/Acer K11/NEC V260X, Canon 60D, macro 60 mm, MagicLantern
Curiousjeff
 
Posts: 230
Joined: Fri Feb 27, 2015 2:17 pm
Location: Geneva, Swizerland

Re: The Great Lines in Scans Experiment.

Postby Curiousjeff » Mon Apr 20, 2015 8:30 pm

I see now that is also a question of luck.

Even with the best camera adjustments, it is not always good.

I guess that it has to do with the start of the video mode on my camera. Sometimes, by luck, it is nearly in sync with the projector. Other times, I see the scrolling lines on the camera monitor and then waves in the scan.

The good thing is I can know ahead of time if the scan will be waveless, but I have not found a way to get it right each time.

Back to your suggestion regarding a precise sync method.

Jeff
Panasonic GH4 4K with lens 12-35, Inogeni 4K to USB3 hdmi capture device, projectors: Acer K135/Acer K11/NEC V260X, Canon 60D, macro 60 mm, MagicLantern
Curiousjeff
 
Posts: 230
Joined: Fri Feb 27, 2015 2:17 pm
Location: Geneva, Swizerland

Re: The Great Lines in Scans Experiment.

Postby Curiousjeff » Mon Apr 20, 2015 8:42 pm

Last try:

Green scan @ Fstop 7 (same as above)

Purple scan @ Fstop 14
Attachments
2015-04-20_21-40-13.png
Panasonic GH4 4K with lens 12-35, Inogeni 4K to USB3 hdmi capture device, projectors: Acer K135/Acer K11/NEC V260X, Canon 60D, macro 60 mm, MagicLantern
Curiousjeff
 
Posts: 230
Joined: Fri Feb 27, 2015 2:17 pm
Location: Geneva, Swizerland

Next

Return to Laser and Camera Hardware

Who is online

Users browsing this forum: No registered users and 1 guest

cron