Sunrise at Coffs Harbour Jetty (teaser) and HDR discussion
Hi guys.
I'm pretty excited while going through some of my shots from Coffs Harbour - there's hopefully some good stuff in here. Here's one of the first from the sunrise session at the Coffs Harbour Jetty.
Trying to reduce 66 photos down to a manageable number that I want to post, is going to be difficult
This image is the result of using the HDR technique, except at the time I didn't take 2 exposures. I processed the RAW twice, once for the highlights and once for the shadows, and blended them in photoshop.
HDR = High Dynamic Range. Your eye can see a lot more tonal range than what can be captured with a camera. Usually a sacrifice is made to either blow out the highlights to get some shadow detail, or sacrifice shadow detail in order to preserve the highlights. But our eye can see the whole scene.
HDR processing involves taking multiple exposures - at it's simplest form, 2 exposures - one exposure for the highlights (where shadow detail is lost), and one exposure for the shadows (where highlights are blown out). The two exposures are then blended together so you get the whole dynamic range in your image.
Quote:
Originally Posted by sheeny
How does this processing compare with the highlight/shadow adjustment in PS CS2?
The shadow/highlight feature really just compresses the histogram - it can't recover data that wasn't there in the first place. If you sky was overexposed because your camera metered on the foreground, you can't recover that data.
By taking 2 exposures, where each exposure properly captures the detail, you get a full histogram with detail preserved at either end.
Just to further clarify, I've used it in a couple of ways on different images:
1. Taking 2 exposures, with the camera on the tripod. Aperture stays the same, ISO stays the same, just change the exposure time. Check the histogram after each shot to ensure you've captured the detail you want.
2. For handheld shots, or where you've only taken 1 shot, if you take it in RAW mode, the 16-bit data allows you to convert the RAW to TIFF, once where you've adjusted exposure for the highlights, and again for the shadows.
The 2 TIF's are then blended in photoshop as per above.
As an example of #2 above (single shot, processed in 2 ways), the attached examples demonstrate it.
The shot was taken during mid-morning with bright sunshine, lots of contrast. I used program auto mode (P), with fill flash, and the camera metered on the tractor, so it overexposed the sky.
The first attachment shows how the shot came out of the camera.
But since it was taken in RAW, there was still some data able to be recovered. Using my RAW editing program (Raw Shooter Premium), I reduced the exposure and curves so that the sky had detail. You can see though that the tractor became underexposed (attachment 2).
The 3rd attachment is the resulting image after blending the 2 images in photoshop. For this one, I copied the the bright image on top of the dark image as a layer, and created a clipping mask. I used the magic selection tool to select all the areas of the sky, and then pressed delete, to reveal the proeprly exposed sky from the layer beneath.
Yeah Mike, I understand the benefit of HDR processing for multiple shots so you can capture the HDR in the first place. No question about the procedure for that.
My question was purely relating to using the one raw image to create to images which you then blend. The raw image has all the data that you have to play with, and the Shadows/Highlights adjustment in PS CS2 lets you do exactly what you appear to be doing by splitting the image and using the HDR blending.
I was just wondering if you had a reason for doing it that way, or if you'd done the comparison to see if there's an advantage.
Good question Al, i'm not sure of the exact reason why - but there is a definite difference.
I went back to the original RAW file of the tractor, and without any adjustments, converted it to TIF and opened it in PS.
Using shadow/highlight, I couldn't recover ANY of the sky detail - because it was clipped - the pixel values showed 255/255/255. There was nothing to recover.
It showed pure white in Raw Shooter Premium as well, but once I reduced the exposure value in RSP, the detail came out (see attachment).
I think it's because once it's converted to TIF, if it was white in RSP, then it's white in the TIF (clipped).
If I opened the RAW in Photoshop, and adjusted it there, I could probably recover the data in the same way.
It sounds right because in PS CS2 when working with HDR it's all handled in 32 bit mode so the range is far bigger than even the monitor can display.
Once it's converted back to 8 or 16 bit tiff the detail is exposed.
If you're using Raw you're still dealing with 16 bit (12 bit actually).
If I opened the RAW in Photoshop, and adjusted it there, I could probably recover the data in the same way.
That's what I was thinking/assuming!
As RB mentioned, a RAW file is 12 bit data, so if you open it in PS you should be able to retain all the 12 bit data that's there. But then you should be able to do that anyway if you converted it from raw to 16bit TIFF as long as you didn't push it too far one way or the other (bright or dark).
Most of the image data sits in the top quarter (highlight end) of the histogram.
Overexposing your image so that the histogram sits towards the right will allow you to recover much more data later (in RAW mode) than having the data spread evenly across the histogram in the first place.
Sunrise at Coffs Harbour Jetty (teaser) and HDR discussion
