[johann.dugge]

Hey there,

thought I'd share this with you lot. It's part of my final year project at uni to develop an HDR camera.

I've only used a matte backing film from an OHP transparency, which gives pretty good resolution of about 1MP but incredibly shitty fall-off, and a piece of 90gsm tracing paper. That one produces much better fall-off, but resolution is impaired severely, I assume due to scattering through the paper. Also needs about 7 extra stops of exposure.

I'm sourcing and will hopefully pick up some frosted glass tomorrow:
http://www.edmundoptics.com/onlinecatal ... uctid=1935

There are also some special films available that would be better suited, but I needed to get this prototype working rather urgently...
http://www.anchoroptics.com/catalog/product.cfm?id=311

The setup:


Sample image with backing film:


Backing film fall-off:


Sample image with tracing paper:


Tracing paper fall-off:


EDIT: added fall-off graphs.

[bjorn.syse]

Sorry, but what are you doing again?

[johann.dugge]

Hey Björn,

I study Product Design and Manufacture.

The idea of the project is to modifiy a camera to capture a higher dynamic range in a trade-off for image resolution - I'm following and implementing a 2000 paper ""High Dynamic Range Imaging: Spatially Varying Pixel Exposures" --> http://www1.cs.columbia.edu/CAVE/public ... PR00_2.pdf

It's a lot more challenging than I thought it would be when I proposed it, but the camera now is a big step forward!

[bjorn.syse]

Ah, now I get it. Cool and good luck!

[johann.dugge]

I worked on the camera setup a bit more, now using Minolta extension tubes so that I can actually change lenses without more glueing. Here's a little sample movie:
http://blip.tv/play/AemUKwA