Page 1 of 1

32-Bit Images

PostPosted: Fri Apr 15, 2005 7:40 am
by SoCal Steve
A friend sent me this link today to a company that sells a program for combining multiple exposures to increase dynamic range to 32 bits. I understand that the new version of Photoshop CS will also do something similar. This may help to explain some of the supernatural effects that we see from the professionals in their images and may give you some ideas for your PP (without having to purchase anything).:wink:

Apologies to those who have seen this before.

http://www.hdrsoft.com/examples.html

PostPosted: Mon Apr 18, 2005 10:21 pm
by sheepie
Had a play with this tonight - looks promising, but also looks like I need more work to get it right :)

Not sure if this offers a huge amount more than PS, will have to consult a couple of the PS guru's around here for some technique polishing! (MattK - you listening?).

Anyway, I put together a quick comparison, using three adjusted versions from the one NEF file (with exposure control adjusted between them). Shown here is the mid-range NEF image, along with the output from Photomatrix:

Image
http://sheepie.smugmug.com/gallery/246573/1/19940675/Large

I am not at all happy with how it has greyed the sky, but that is probably more to do with my settings on it than anything else!

I'll do some more experimenting and feed back down the track :)


Thanks for this link Steve.

PostPosted: Mon Apr 18, 2005 10:32 pm
by kipper
Looks pretty good. I'll have to dig out some of the river shots now :)

PostPosted: Mon Apr 18, 2005 10:37 pm
by Matt. K
Sheepie
I presume you combined 3 layers to do this? It should be possible to rub through the top layers with the erasor to knock out the offending tones.
By the way...the rest of the image is supurb!

PostPosted: Mon Apr 18, 2005 10:40 pm
by kipper
I dunno, I think the issue is that the highlights look blown in the first image?

PostPosted: Mon Apr 18, 2005 10:43 pm
by kipper
Just checked out the program website. You'll notice that the darkest image in the examples have the sky with no blown highlights.

PostPosted: Mon Apr 18, 2005 10:54 pm
by sirhc55
The problem may have an answer in this:

A 32 bit image consists of 3 color planes, usually red, green and blue, and an additional alpha (or transparency) plane.

Imagine our image consists of a color ramp, left to right, from black to red. Using the 8 bits available for red in a 32-bit system, we can have at most 256 levels. Therefore, if our image is greater than 256 pixels wide, we have to duplicate red values. While this is not so noticeable at standard video resolutions, HDTV, photographs or images at film resolution (which are generally about 4000 pixels wide, but it can be as high as 16000 pixels for IMAX) will have large areas of red that are all the same value. This can be quite noticeable because the human eye is extremely efficient at picking out edges. This imaging artefact is known as banding, and can also be noticed in the alpha channel when compositing images together.

PostPosted: Mon Apr 18, 2005 10:55 pm
by sheepie
kipper wrote:You'll notice that the darkest image in the examples have the sky with no blown highlights.

Which I think is my problem here. I'll have to try some without any blown highights and see what happens - seems to cope well with the rest of the pic, although as Matt says, you can probably achieve the same effect in PS (albeit with a little extra work!).
I usually make a point of ensuring no blown highlights, but every now and then there is no choice! I guess I should start bracketing some of these shots - better to have more files to process than to loose a shot :)

PostPosted: Mon Apr 18, 2005 10:58 pm
by sheepie
sirhc55 wrote:The problem may have an answer in this:

A 32 bit image consists of 3 color planes, usually red, green and blue, and an additional alpha (or transparency) plane.

Imagine our image consists of a color ramp, left to right, from black to red. Using the 8 bits available for red in a 32-bit system, we can have at most 256 levels. Therefore, if our image is greater than 256 pixels wide, we have to duplicate red values. While this is not so noticeable at standard video resolutions, HDTV, photographs or images at film resolution (which are generally about 4000 pixels wide, but it can be as high as 16000 pixels for IMAX) will have large areas of red that are all the same value. This can be quite noticeable because the human eye is extremely efficient at picking out edges. This imaging artefact is known as banding, and can also be noticed in the alpha channel when compositing images together.


...and all this at 10pm! It seems to make some sense, but I think I'll have to have another look at this one tomorrow :wink: