Integrams?

Have a great new article on a science related topic? Post it here!
Colin Kaminski

Integrams?

Post by Colin Kaminski »

I have the integral papers linked on the lippmann section of the wiki. I am afraid they are not in English but translations will come in time....
WingedOne

Integrams?

Post by WingedOne »

Those links aren't working. :(
JohnFP

Integrams?

Post by JohnFP »

WingedOne, This is very interesting. But I am not sure if I follow your technique. Do you render the image in POVRay and then print it out on a transparancy then contact copy it? Or do you have a lens array that you somehow matched to the POVRay rendition? Do you mind explaining?



Also, what depth have you been able to achieve so far and what emulsion are you using?
Colin Kaminski

Integrams?

Post by Colin Kaminski »

WingedOne wrote:Those links aren't working. :(


:oops:



Ok, I fixed the first one, for the second one, I need a different computer so I have to get to it in a day or two.



In the mean time you can look for it in this list:



http://www.holographyforum.org/lippmann/



This link will work until I get it fixed:



http://www.holographyforum.org/lippmann ... 2_1908.pdf



It is the %20 that is the problem so I will rename it to have an underscore when I get to that computer.
WingedOne

Integrams?

Post by WingedOne »

JohnFP wrote:WingedOne, This is very interesting. But I am not sure if I follow your technique. Do you render the image in POVRay and then print it out on a transparancy then contact copy it? Or do you have a lens array that you somehow matched to the POVRay rendition? Do you mind explaining?

Also, what depth have you been able to achieve so far and what emulsion are you using?


No emulsion was used. It might have been a bit confusing with the other article that I referenced in my first post. I just posted the reference to the article because I was wondering if anyone else here had read it and tried that technique with photography-based integrams.



I basically set up a computer-generated array of concave lenses within POV-Ray between the camera and the subject matter. In pov-ray I used an orthographic camera.



I rendered the image at a resolution of 4000 x 3200 pixels and printed out the image on a Lexmark photoprinter, cut off the borders of the print and placed it in the frame behind the #300 fresnel lens array.



The printout was on Hammermill 110 lb card stock since it's less prone to distortion and I get a cleaner image. When I tried printing on photo printout paper I got depressions in the paper in the darkest areas because of the ink and when I placed it behind the lens array, I got colored halos around the objects in the scene.



I can get a decent depth of about 3 to 4 inches. My printer has a dpi of 4800. I figure if I had a printer of 9600 dpi, I might be able to get better depth.



I tried rendering at 8000 x 6400 pixels but it caused my computer to crash when I tried to print it. :?



I'm rendering at a slightly higher resolution of 5000x4000 pixels right now and I'll see if that gives me any better results.
JohnFP

Integrams?

Post by JohnFP »

That is simply fantastic. Thanks!



So, you need to keep the lens array in front of the rendered image to view it, correct?



Then I belive you mentioned making a hologram of that image....is that the idea?



I have been working on something similar and so far it has been just thought experiements and some trials with a small single lens (1/4"diam) which would become an array in the final and/or a cylindrical lens which would become a series in the final.



My idea was to take sucessive shots of an image. Take each one of these shots and shrink them to the size of the lens array or cylindrical series. Then have them all output to an LCD screen at the same time. Take the lens array or cylindrical series and place it in front of the "rendered" images and have them focus back to overlapping areas in space. Place a holographic plate there and create a one step stereogram.



Now, the technique you describe makes this easier. The key would be to get an LCD tranmission screen that has a very high resolution. Mine is only 800 by 600 pixels.



Let me ask you a question. Is a real image focused in front of the lens array? That is, can you light the image and lens array brightly in a dark room and see a focused image in front of the lens array somewhere? You may need another giant fresnel lens in front of the array to image it.



This is wild, I like it.
LarryD

Integrams?

Post by LarryD »

Here's some historical info on these techniques.



http://www.microlens.com/HistoryofLenticular.pdf



Theres also an interesting blurb on the Kodak Kodacolor process in there.
WingedOne

Integrams?

Post by WingedOne »

If there is a real image I think it would be behind the plane of the lens array. Sometime, I'll have to try printing an image out on a transparency and try it out.

On a related note, I just came across this insteresting article where some researchers created a holographic lens array for integral imaging which might lessen some of the disadvantages of conventional lens arrays.

http://spie.org/x8756.xml?ArticleID=x8756

Is there any information anywhere on how to make holographic lenses? :think:
MfA

Integrams?

Post by MfA »

I don't see why you couldn't get a real image in front of the lens array.

As for making holographic lenses all you need is access to a sub micron greyscale laser plotter or direct write lithography system :) (They probably used an e-beam writer for the HOE in that article.) The rest is trivial, you just print out an array of sinusoidal zone plates. All it would do is increase the viewing angle though, you would still need ridiculous resolution images to get decent reproduction. Integral images are not so different from holograms in that respect.
favalora

Integrams?

Post by favalora »

Hi -

I just wanted to say that this is interesting work, and I enjoy reading about it on this forum. There is considerable research on integral photography in Asia, and very little here in the States.

Did you go through an annoying process of matching your printer's resolution to the exact pitch of the lenses? Registration is a difficult matter for these things.

Although it may help for you to increase the dpi to get more depth, keep in mind that commercially-available lenticulars and fly's-eye lens arrays suffer from significant aberrations. Furthermore, you will hit a diffraction limit. I can post links to the relevant papers if you want. Some very good work was done by Stephen Daniell of Zograph but he had to halt his work. Anteryon (??) within Philips's building also was capable of good quality lenticulars. At one point we commissioned them to make doublets for us at fine pitch....

I never did the raytrace myself, but I am curious if you're actually getting a pseudoscopic image. I think you have to do some careful processing if you want the imagery to come out the right way.

Can you post a movie?

Keep it up,
Gregg
Locked