Daniel Reetz, the founder of the DIY Book Scanner community, has recently started making videos of prototyping and shop tips. If you are tinkering with a book scanner (or any other project) in your home shop, these tips will come in handy. https://www.youtube.com/channel/UCn0gq8 ... g_8K1nfInQ

Sparkles on DIY Hackerspace Scanner

All about lighting. LED, CFL, Halogen, Other? Questions and info about lighting go here.
Post Reply
Javier_Romagosa
Posts: 3
Joined: 14 Aug 2012, 04:31
E-book readers owned: Pantech Element LP4800
Number of books owned: 1
Country: Spain

Sparkles on DIY Hackerspace Scanner

Post by Javier_Romagosa » 03 Dec 2012, 07:15

Good morning,


I´ve been successfully scanning books with my Hackerpace Scanner for nearly a month. However, when I have attempted to do the same with a magazine two sparcles appear in scene and spoil the work.

Some people have advised me to buy sheet polarizers but, since they are pretty expensive I've thought of asking the community first. So, what do you think of it? Any other solution comes to your mind?


Thanks in advance,



J. Romagosa


Image

Uploaded with ImageShack.us

dpc
Posts: 275
Joined: 01 Apr 2011, 18:05
Number of books owned: 0
Location: Issaquah, WA

Re: Sparkles on DIY Hackerspace Scanner

Post by dpc » 03 Dec 2012, 13:38

What you're seeing is the reflection of the light off the platen and back into the camera's lens. Typical mitigation strategies are to move the light higher, or move the camera farther away from the platen and zoom in (i.e. decrease the camera's angle of view), the former being the easiest solution with a scanner like yours that is already assembled.

BTW, I can also see two vertical brown bars in that photo you posted. From another post that you made today that includes a picture of your scanner, you've chosen to stain the wood instead of painting it flat black. So it appears that you're getting reflected light off of the scanner's sides being reflected off the platen as well. The stained wood does look nice but perhaps you could paint just the surfaces that are causing these reflections flat black to eliminate this artifact (or cover the edges with flat black tape?).

Tip: When chasing down reflection/glare problems, put a piece of solid black paper that is the dimensions of the largest page you're planning to support under the platen in place of a book or magazine. The black paper will reveal any reflection problems easier than what you'd see with the typical printed page.

LA2
Posts: 19
Joined: 01 Feb 2012, 16:43
Number of books owned: 1000
Country: Sweden
Location: Linköping, Sweden

Re: Sparkles on DIY Hackerspace Scanner

Post by LA2 » 09 Dec 2012, 22:04

There is a forbidden zone, where lights should not be because they will cause reflexions. This zone is a rectangular cone that starts at the camera lens and goes to each corner of the glass, where it is mirrored, and forms a bucket shape around the camera. At the camera, this zone is always twice as wide (2w) as the glass. If the camera is close to the glass, like the Ion Audio Booksaver, the cone or bucket will have an open angle and there is no way to combine a V-shaped glass, because the two buckets will overlap, leaving no place for any light at all. To avoid reflexions, any light source needs to be placed between the two buckets. For this, you need more narrow cones, which requires a larger distance between the camera and the glass.
bucket2.png
Forbidden zone, or bucket
bucket2.png (18.77 KiB) Viewed 5308 times

User avatar
jck57
Posts: 376
Joined: 23 Nov 2009, 15:21

Re: Sparkles on DIY Hackerspace Scanner

Post by jck57 » 11 Dec 2012, 08:37

LA2 wrote:There is a forbidden zone, where lights should not be because they will cause reflexions. This zone is a rectangular cone that starts at the camera lens and goes to each corner of the glass, where it is mirrored, and forms a bucket shape around the camera. At the camera, this zone is always twice as wide (2w) as the glass. If the camera is close to the glass, like the Ion Audio Booksaver, the cone or bucket will have an open angle and there is no way to combine a V-shaped glass, because the two buckets will overlap, leaving no place for any light at all. To avoid reflexions, any light source needs to be placed between the two buckets. For this, you need more narrow cones, which requires a larger distance between the camera and the glass.
bucket2.png
Thank you for this essential, simple explanation.

Javier_Romagosa
Posts: 3
Joined: 14 Aug 2012, 04:31
E-book readers owned: Pantech Element LP4800
Number of books owned: 1
Country: Spain

Re: Sparkles on DIY Hackerspace Scanner

Post by Javier_Romagosa » 12 Dec 2012, 08:17

I understand. Thank you very much.

Since it seems pretty easy, I'm going to elevate the light to avoid the forbidden zone. Also, I'll paint in matt black the parts of the chasis that are causing trouble (thank you dpc for that, I hadn't notice at all).

Best regards from Spain,



Javier Romagosa

kalyan
Posts: 17
Joined: 11 Sep 2012, 02:51
Number of books owned: 0
Country: India

Re: Sparkles on DIY Hackerspace Scanner

Post by kalyan » 13 Dec 2012, 04:11

You may position the LEDs as per the image to avoid reflection.
Attachments
light position.png
light position.png (10.2 KiB) Viewed 5229 times

dpc
Posts: 275
Joined: 01 Apr 2011, 18:05
Number of books owned: 0
Location: Issaquah, WA

Re: Sparkles on DIY Hackerspace Scanner

Post by dpc » 13 Dec 2012, 14:38

Technically, you're correct. You could also move the lights fore and aft to reduce the reflections back into the camera. There's an important bit you might have overlooked though. The camera's viewing angle is wider in that direction (camera aspect ration is wider than it is tall) so you'd have to move the light at a lower angle to the platen surface in order for it to not show up as glare. That would reduce the overall lighting intensity at the platen and make things more difficult to avoid shadows cast from the operator.

Not to mention that with most of the scanners built by folks on this site, the front and back of the scanner have frames or cross members that would project a shadow on the platen from lights positioned fore/aft.

Another way to visualize what LA2 demonstrated with his diagram is to imagine that the camera position has been flipped around on the other side of the glass and is now pointing up through the glass. Anything that is in the camera's view from that position will be reflected back on to the glass surface when the camera is at its normal position.
bucket3.png
camera reflected about the platen surface
bucket3.png (32.19 KiB) Viewed 5218 times

Post Reply

Who is online

Users browsing this forum: No registered users and 2 guests