Scanning Objects

[G. Couch]

This may have been discussed before, but I could not find an older thread in my search.

Anyway, I noticed some of Gerry Monaghan's scanned flowers in the gallery and I posted some images I have made using scanned shells. Does anyone else scan objects like flowers, leaves....bugs?

Does anyone know of any good tips or sites that discuss this? I have been using a couple methods for my scans of shells. One is to use a box with the inside painted black, placed over the object. This provides a nice, solid black background around the object and makes it easy to extract it from the background. I have also used a box with a bright white interior as a way to reflect some of the scanner light back down onto the object.

If anyone has the latest National Geographic, there is an excellent article on moths and all of the moth photos were created using a flatbed scanner. The resulting images are breathtaking in their detail.


I have attached one of my seashell scans. The shell measured just over 1 inch (25 millimeters).

[Ed_L]

No words of wisdom here Greg. But it looks like you're doing some things right! Looks great!

Ed

[Sanda]

that's great Greg, it looks more like a photo than as scan. Looks like you don't ned any help you've got it working well.

[jerry]

Hi Greg

That is really interesting. I had no idea that you could scan objects with such clarity. I had built a small tent type device with a reflective surface to try to scan slides. I had no success the slide was always very dark. The moth idea sounds very interesting. I will watch for posts.

Thanks

Jerry

[DJ Dubovsky]

What do you need a tutorial for? From the looks of that shell you could write your own tutorial. That turned out very crisp. Excellent job. I will give your black and white box a try. I like the results and your are right, it does make it easy to remove from the background.
DJ

[G. Couch]

DJ, Sanda, Ed - Thanks for the nice compliments! I have been pretty happy with the results I am getting with my new scanner, but my old scanner was a nightmare when it came to scanning objects. Things had to be perfectly flat in order to get a sharp scan. I think flatbeds that have transparency capabilities have a greater depth of field than ones without. Does anyone have any technical knowledge about how to "trick" a scanner into increasing the depth of field?

Jerry - I built one of those back when I had my first scanner (rest in peace poor Umax...). Like you, I had very little success! You should check out the National Geographic article. They caught literally hundreds of different species in 1 ordinary backyard. They scanned them all on a high-end flatbed and made 34"x48" prints. If I can catch a moth or a butterfly I will attempt to scan it and post it here...I'm just such a wimp when it comes to killing even little bugs.

[OhThatGirl2001]

Hi Greg,

I have tried occasionally to scan a 3-dimensional object with my HP scanjet 5470C. The results were not bad considering it was a flatbed scanner. I used a flat black piece of paper and mounted it on the top of the scanner. You might find the following link helpful

http://www.agfanet.com/en/cafe/photo...chapter05.php3

At the bottom are a few newsgroups that I have frequented and found very helpful.

Let me know how it goes or if you want to see the results of my scanner.

Lisa

[Doug Nelson]

Hmm...interesting experiment idea. Mount page full of text at 45-degree angle and put it on a scanner. Scan and examine depth of field. Change a setting, repeat. Repeat, repeat, repeat

Compare with other scanners with the same test (maybe newspaper or something else with standard print). Perhaps even a grid could be made that adjusts for angle and displays distance in mm or inches, providing a standard if printed at 100% (maybe a non-angle-adjusted measurement tool could be included for verifying lack of scaling distortion).

This could be a very original contribution here. Too bad I'm the unmotivated sort

[OhThatGirl2001]

Here is another link you may want to look at.

http://www.stereoscopicscanning.de/

Lisa

[G. Couch]

Lisa - Thanks for the links! You should post some of your images.

Doug - That is a great idea! I decided to do a quick, very unscientific, measurement. Here is the setup I used...it should be repeatable for just about anyone. I was pretty surprised at just how far out the scanner could focus.

oh...I was going to calculate all the angles but it seems that Trigonometry class has faded from my memory...sigh...

[Doug Nelson]

Well, let's see...decent focus up to about the 4" mark, which would be 2", which would be 1/3 of the length, which (keeping angles constant) would make for a field depth of 2/3" (Or would it? My head hurts now).

Math, anyone?

Regardless, it was deeper than I imagined. Now the question is: can you make it deeper on the same scanner?

[Doug Nelson]

Just for fun I looked for links:

Here's one, and here's another. Here's my search results.

[G. Couch]

Good links! I'm not sure if there is a way to increase the depth of field except on very high end flatbeds.

I have read a few articles that say most flatbeds are sharper and have the greatest depth of field at the center of the CCD.

[Doug Nelson]

Oh yeah, I forgot one of my favorite object scanning links
(caution: can be vulgar)

[OhThatGirl2001]

You did your math right Doug!
The break down would be as follows:

2/6 = x/2
2x2 = 6x
4 = 6x
4/6 = 2x
2/3 = x

2/3" is the field of depth.

[G. Couch]

Doug - I remember seeing that site! Poor cats... If I tried that with my cat, I would end up in the hospital with severe lacerations.

Lisa - That makes me feel bad that I can't even remember basic math!

[OhThatGirl2001]

That's OK Greg, my husband who has his Doctorate in electrical engineering was reading over my should and said - Yup, Doug's right. Math was not my best subject.

Lisa

[OhThatGirl2001]

OK Doug,

That's just about the weirdest link I've seen... and I've seen a few weird ones in my time - LOL

Lisa
Meeeowwww

[G. Couch]

Well, I did not manage to get the cat scanned today ...but I did find a interesting feather. I believe it's a hawk feather. Anyway, I figured it would make a good object to scan due to the complexity and detail.

I have to say, I was pretty amazed by the scan! I placed the feather on the glass and put a shallow cardboard box with bright white paper lining the inside. I like the look this gives and I think having some of the scanner light reflect back down helps bring out detail in delicate or translucent objects.

The scan was done at 1200 ppi. This really gives a good idea of just how sharp even a fairly inexpensive flatbed scanner can be.

[jerry]

Wow

That is really amazing. Now I have to start fooling around with scanner boxes. Thanks for the info. Greg.

Jerry

[debelieson]

What a fascinating thread (except for the math parts, of course)! Lots of great information and links. It's got me wondering what objects I can scan, without moving too far from my chair. Thanks.

[G. Couch]

Here is the National Geographic article I mentioned...Moth scans


They used an EverSmart Pro II scanner... a bit more than most of us can afford!

- Greg

[OhThatGirl2001]

You've peeked my interest... I tried scanning a pen do see how well my scanner does with 3-d objects.

Lisa

[OhThatGirl2001]

I guess I'll have to wait until my kitty brings in a feather to see how well it does with that. Did you get a chance to check out the alt.binaries scanners. I found it very useful.

Lisa

[G. Couch]

Lisa - I've been on that newsgroup a few times and found it to very helpful when I was looking to buy my current scanner. I have not been on newsgroups much since finding this site though!

I scanned a few more things today, including a dead cricket , a cool chunk of bark, and a really old key. I'm thinking of devoting a chunk of my web site to a page of scanned objects. I'll post it if I ever get around to making it! (...and if I ever stop making silly album covers )

Scanning objects seems to be a good way to really push your scanner to it's limits. A real object contains far more visual information than even the sharpest transparency. A good example is when I scan an object at 1200 ppi and then again at 2400ppi. When I have done this with slides or negatives, I can only tell a small amount of difference between the two scans. This led me to believe that Epson's claim of 2400ppi optical was a bit exaggerated. However, scanning a real object at the two different resolutions resulted in a tremendous difference!

[Ed_L]

Wow! I haven't seen this thread in a couple of days. You guys have been from rocket science math to scanning dinosaur leg bones! It's all very interesting.

Ed

[G. Couch]

Ed - I'm leaving the math to Lisa and Doug!

I'm not up to dinosaurs yet, but I had fun scanning a cricket...The spiders in the house are quivering in fear!

[G. Couch]

a spider's head...the marks in the picture are millimeters

[OhThatGirl2001]

Kewl. I can see there is no arachniphobia with you! I took a great spider shot close up with my old camera that was kind of freaky looking. You could see the eyes and sack close up.

P.S. Like the new avatar. At least you don't get accused of making a 70's avatar (ha ha BigAl)

Lisa

[OhThatGirl2001]

Here's my little contribution to the spider world