Glare Reduction Techniques

[univurshul]

Not sure if it helps, but I know it doesn't hurt. Couple beers and hot glue gun on a Friday night. Like old style.

[daniel_reetz]

Not sure if it helps, but I know it doesn't hurt. Couple beers and hot glue gun on a Friday night. Like old style.

That's how I do it. I think shades are really important, m'self.

[univurshul]

While this topic is on glare reduction, I've also updated another post on some test-findings of newer plastics and glass w/ anti-glare & anti-mar treatments. Those details are documented in an upcoming build here: http://www.diybookscanner.org/forum/vie ... ?f=1&t=449

If anyone has any important FYI glare solutions or links within the forum, network this thread.

Thanks.

[translucent1]

I'm just finishing up my scanner build, and was surprised by a few quick experiments I tried with lighting. I initially tried mounting two spotlights directly above the platen like most people seem to do, but the glare was terrible. I was getting a very clear reflection of the light bulbs from the camera's point of view. Rather dismayed, I decided to look at the setup used in commercial book scanners.

The Atiz has LED lights mounted behind the platen, facing toward the operator. You can see it in this video clip:
http://www.youtube.com/watch?v=JkutMFVwu0o#t=0m44s
and there's a nice cutaway drawing in the brochure:

LED lighting on the Atiz BookDrive

atiz_lighting.jpg (51.45 KiB) Viewed 9106 times

At the moment I'm setting up an LED spotlight at the rear of my scanner, pointing down at about a 45 degree angle, and there's absolutely no glare, even on glossy pages. It's only a 4W light, but I'm still getting about 1/4s shutter speed at F/7.1 (which is plenty fast, it's not like the book is moving), and that's at +1 EV to give me nice bright white pages. I'm planning on using two of these lights, one for each page.

The downside is that some of the light bounces up into the operator's face, so I don't think I'd recommend this approach with a megawatt of halogen floodlights!

[univurshul]

Atiz® uses an overhead light on the front in addition to the rear--nice. This is right in line with where I'm going. When you look at their distances from the platen surface and the platen angle, it's not conducive to a direct overhead lamp, you can see why they line it up that way--it's otherwise too close. Simply less light-bounce toward the camera optics when the bulbs face in parallel with the platen seam.

Thanks for the post.

[univurshul]

This is perhaps one fail-safe method: black-out one side of the platen.

This obviously cuts workflow literally in half. Having said that, I've already scanned this book before and I'm going back to scan it again because of glare issues.

Why will it take twice as long to scan this way?: Because one side of the Platen will be blacked-out with a light-absorbing material. I suppose you could alternate setting the mat paper on each side of the platen as you go, but it would slow you down even further (scan, process, data, etc.), plus there is a bit of technique setting the mat paper very close onto the pane: double-side sticky tape; not too much.

Not every book I scan is "slash & trash", not every book is a binary processing achievement (text-body scanning). I'm talking about scanning visually stunning books, design books, architectural books. This means I'm after the best DIY reproduction possible, and if that means I need to go through the book 2x, so be it. It's cleaner reproduction than air-platening, and I believe some of the reflective glare off these glossy pages are inhibited by having an optic-pane resting over it during capture.

Until I find platen material that cuts glare while maintaining transmission/clarity benchmarks...

[univurshul]

I'm just finishing up my scanner build, and was surprised by a few quick experiments I tried with lighting. I initially tried mounting two spotlights directly above the platen like most people seem to do, but the glare was terrible. I was getting a very clear reflection of the light bulbs from the camera's point of view. Rather dismayed, I decided to look at the setup used in commercial book scanners...At the moment I'm setting up an LED spotlight at the rear of my scanner, pointing down at about a 45 degree angle, and there's absolutely no glare, even on glossy pages.

What kind of material and thickness is your platen made of?

--Thanks

[translucent1]

What kind of material and thickness is your platen made of?

--Thanks

I'm using 3/16" scratch resistant cast acrylic. I've forgotten the exact brand name, but it's something like this:
http://www.techplastcoated.com/info-opt ... cratch.asp

[univurshul]

nice. Can't wait to see what you're building.

[univurshul]

At the moment I'm setting up an LED spotlight at the rear of my scanner, pointing down at about a 45 degree angle, and there's absolutely no glare, even on glossy pages. It's only a 4W light, but I'm still getting about 1/4s shutter speed at F/7.1 (which is plenty fast, it's not like the book is moving), and that's at +1 EV to give me nice bright white pages. I'm planning on using two of these lights, one for each page.

I just completed a test run of 1 of my 1900 lumen-output halogens directed at a 45 degree angle.

Results: same glare issues mounted at 45 front/rear compared to 45 mounted from overhead.

IMG_0002.jpg (34.74 KiB) Viewed 9096 times

Not good. it's time to reduce lumen, spot-intensity. The lights directed at a 45 degree angle still might be a great solution, but my light source need to "wash" over the platen as opposed to blow it out with an inferno sun ball

Which leads me to some important questions: If I simply lower my lumen output with wide-angle cascading lights, will this solve a lot of glare issues? Are lumen intensities/distances directly correlated to the amount of resulting glare on my platen surface?

I'm discovering It has more to do with the lumen output/distance hitting the surface. It has more to do with the coating, the finish, composition and angle of the platen. It has to do with spot-instensity, lumen count, lamp distance.

It has less to do with placement of lights, although you can easily make it worse if they're setup in the most inopportune spot.

It's only a 4W light, but I'm still getting about 1/4s shutter speed at F/7.1 (which is plenty fast, it's not like the book is moving), and that's at +1 EV to give me nice bright white pages. I'm planning on using two of these lights, one for each page.

I have a feeling when you crank up the wattage on your LEDs you'll need to watch out for how bright, how narrow the light beam is. I just tested PAR30 LED halogens and they are 9 watt bulbs. My total lumen output with these 2 LEDs is 1080lm. Glare was less than my 3800lm-output halogen incandescents, but the light didn't wash evenly over the platen.

IMG_0001.jpg (32.67 KiB) Viewed 9096 times

Atiz® is using LED fixtures that diffuse well around the booth, it's not a directed bulb persay, yet strong enough to illuminate the platen. I wonder about their platen material, angle...

Some of my already scanned books I can live with, but the favorites I captured with intent for digital preservation need a new treatment.

Most glaring is subtle. When you look at my examples in a prior post you'll see the reflectivity along the platen seam is minor, but once addressed with an opposing black page, it makes a world of difference.

I'd like to see how your shots come out on large art-print books and media around 11"X13"/page.--That is a worthy test of the majority of the platen's real estate and it's relationship to the lighting.