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

building a motorized scanner

building a motorized scanner

Postby jbeale » Sat Aug 18, 2007 7:56 am

I want to see if I can get a better scan if the laser is moved very smoothly at a constant speed. If anyone else wants to try doing this, I found a nice gearmotor from www.allelectronics.com for only $9.
http://www.allelectronics.com/cgi-bin/i ... 2VDC_.html

This item is listed at 5 rpm at 12V dc, but the two I just got still run smoothly down to 1.6 V input, where the output shaft turns at 0.2 RPM. That means it takes about 47 seconds to sweep out a 90 degree angle.

I haven't used it for a scan yet, but I just rested my laser level on the output shaft and watched it sweep slowly around the room. Very interesting to watch it trace out all the contours- almost worth the price just to see that.
Attachments
DCM-276.mov
video showing gear train inside All Electronics DCM-276 motor with output shaft (bottom black gear) at 0.2 RPM
(815.79 KiB) Downloaded 1534 times
scan1.mov
slow laser sweep across desk
(471.95 KiB) Downloaded 1409 times
jbeale
 
Posts: 41
Joined: Sat Aug 11, 2007 9:20 pm

actually about 0.33 rpm

Postby jbeale » Sat Aug 18, 2007 4:08 pm

ok, my math was wrong- hey, it was late. I measured 90 degrees in roughly 45 seconds, which corresponds to 0.33 rpm. At any rate, it is nice and slow.
jbeale
 
Posts: 41
Joined: Sat Aug 11, 2007 9:20 pm

Postby WalterMo » Sat Aug 18, 2007 9:01 pm

When I was looking for a very slowly DC motor with gear I could not find such one and decided to build up a scanner with a stepping motor.
The same motor as presented by jbeale you can buy now in Germany from www.pollin.de for 5.95€.
It's called there Gleichstrom-Getriebemotor 04885206.
User avatar
WalterMo
Moderator
 
Posts: 2363
Joined: Mon Apr 02, 2007 6:52 pm
Location: Braunschweig, Germany

Postby MagWeb » Sat Aug 18, 2007 10:28 pm

Ok - this is not a revolutionary idea:

Remember your toys.

Perhaps you own a toy-construktion-system like "Mecano", "Eitech", "Fischertechnik" or something similar, sleeping in your cellar.

With my 30-year-old Fischertechnik-parts I reduced the DC motor to about 0,25rpm building a simple transmission to rotate the laser. A very variable possibility for testing.

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

another plea for unweighted cumulative average

Postby jbeale » Mon Aug 20, 2007 8:59 pm

I made a "barn-door" type mount for my laser with two pieces of wood, a hinge, and all-thread shaft + nut. This is a common camera mount for amateur astronomy, for example http://www.zodiaclight.com/equipment/barnDoor.htm

I didn't calculate it out before hand -oops- so I now realize with my current motor drive, my laser sweep rate is from 2 deg/min adjustable down to 0.12 deg/minute. The slowest rate is about half the speed that the sun moves across the sky! The fastest is still only 3 rev per hour. I need a faster motor.

But meanwhile, it would be *really nice* if DAVID allowed unweighted averaging, so at least I would get the full noise-averaging benefit of the slow scan. As it is now, when the latest sweep forms 70% of the total, there is very little cumulative averaging (= noise reduction) over time. I think adding this as an option is a fairly easy software change, yes?

if Sample(n) is the nth sample you have acquired and Data is the saved result, then

Old way: New_Data = 0.3 * Data + 0.7 * Sample(n)
New way: New_Data = ((n-1)/n) * Data + (1/n) * Sample(n)

the only new memory requirement is an integer array of the "n" values for each image pixel, which is the number of times that pixel has been sampled before.

As a side-note, the ability to display or export that "n" value array might be of some interest, as well.
jbeale
 
Posts: 41
Joined: Sat Aug 11, 2007 9:20 pm

spinning mirror scanner

Postby jbeale » Sun Sep 02, 2007 11:19 pm

I added a spinning polygonal mirror to my "barn door" type tilting platform to try to improve my laser line quality. It works and I am happy with the uniform narrow line I am getting, but I have not tried a 3D scan with it yet so I don't know if it is really better. I have included a photo of the scanner, and the beam hitting my arm and hand.
Attachments
laser-scanner.jpg
spinning mirror type laser scanner on tilting platform
laser-line-arm.jpg
laser line from spinning mirror scanner on my arm & hand
jbeale
 
Posts: 41
Joined: Sat Aug 11, 2007 9:20 pm

results from rotating mirror scanner

Postby jbeale » Mon Sep 03, 2007 3:10 am

Here is a test scan of a plastic water pitcher, about 8 inches high. The surface should be perfectly smooth but the model still looks a little rough. This is after one step of median smoothing in David, saved out to OBJ, viewing in MeshLab. I don't know what I'm doing wrong; it seems to me like the surface should be smoother still.
Attachments
water-pitcher.jpg
one view of a 8" plastic water pitcher, using rotating-mirror scanner
jbeale
 
Posts: 41
Joined: Sat Aug 11, 2007 9:20 pm

Postby Bongobat » Tue Sep 04, 2007 1:07 am

What would you estimate the thickness of your line is? What camera are you using? Did you scan in Black and White or color? What is David showing as your FPS while you scan? Can you post a picture of the raw scan. What version of David are you using? Basically give more info on your specs.
User avatar
Bongobat
Moderator
 
Posts: 1002
Joined: Mon Apr 02, 2007 10:02 am
Location: Ohio, USA

Postby jbeale » Wed Sep 05, 2007 9:18 pm

The laser line thickness is somewhere around 1 mm, depending on how bright you set the camera exposure. I'm using a Sony FX1 (1080i HDV) camera set to black&white and output DV (720x480). In the black & white mode, the DV raster does have 720x480 resolution even with the red laser line (which would be blocky in color mode due to DV chroma subsampling.) David version 1.5 reports 29 fps. The video image looked pretty clean but sorry I don't have a frame capture right now. Actually I've given up on this for now because I was not satisfied with the output after trying several different things. Maybe sometime I'll come back to it.
jbeale
 
Posts: 41
Joined: Sat Aug 11, 2007 9:20 pm

Stay with it!

Postby Bongobat » Tue Sep 11, 2007 8:06 am

I meant in reality what is the size of you line. How does that camera interface with the computer? USB, Firewire, or through a capture card? For God's sake dont give up! When you find the magic settings youll be amazed how cool scanning is. I have gotten excellent results with a fairly cheap USB webcam (spc900nc) and a laser pointer with a diffractive line generator fitted to the front. I project a line about 1mm thick at most, but thats just an estimate from eyeing it. A high resolution camera is only usefull if your line is very very thin. And those cameras are hard on the computer causing low fps which make it hard to scan. My cam gets around 30 fps at 640x480 and around 70 fps at 320x240. I usually scan at 640X480 which creates a fairly large polygon count. I then use nevercenter's Silo and there topology tool to recontsruct it to a useable size for animation. Oh and if Simon and Sven are listening you made version 1.5 too good and now you need a slider or something on the smooth average tool so you can set it to a lesser power. I hardly even have to smooth now.
User avatar
Bongobat
Moderator
 
Posts: 1002
Joined: Mon Apr 02, 2007 10:02 am
Location: Ohio, USA

Re: building a motorized scanner

Postby jonb2000 » Wed May 06, 2009 4:46 pm

Hi jbeale,

I recently bought the gear motor you suggested in your last e-mail but I got the motor without the power source.
Can you please let me know where can I get the power source for the motor.

( gearmotor from www.allelectronics.com for only $9. http://www.allelectronics.com/cgi-bin/i ... 2VDC_.html)

Thanks,

jonb2000
jonb2000
 
Posts: 14
Joined: Mon Mar 16, 2009 3:35 pm

Re: building a motorized scanner

Postby Khalid Khattak » Wed May 06, 2009 4:57 pm

I think your laser setup is well organized..i liked it :) .. As far as the quality scan is concerned i am still after it.. i will be happy in the day when i scan a coin?... well... can we scan the coin with good details with the help of Logitech 9000 and a focusable red laser????
PC: Dell Inspiron i3, 2.27GHz,4GB RAM,64 bit windows 7
CAMERA:
David CCD Mono USB
LASER:
David 5mW Green line laser
PROJECTOR:
3D LED Mini Projector 250lumens, Native 1024x600
[b]http://free3dscans.blogspot.com/
Khalid Khattak
 
Posts: 942
Joined: Thu Apr 23, 2009 1:30 pm
Location: Pakistan


Return to Hardware

Who is online

Users browsing this forum: No registered users and 1 guest

cron