The answer I will be looking for in this matter is probably well known to those who use RAW processing, but I don't yet, so I'm looking for quick enlightenment!
I have assumed that the parameters of the white balance adjustment applied to an image during in-camera processing are somehow saved in the resulting JPG file (perhaps in the "proprietary" portion of the Exif metadata, like the focusing point selection info).
I make this assumption because, when we go to make a custom WB setting, we do it by taking a shot (ideally, one whose color balance reflects the balance of the ambient light we are interested in, by using a neutral test card, an ExpoDisc, or a wad of coffee filters. We then nominate this image to be the basis for the establishment of our custom set of WB parameters.
But WB correction has already been applied to the image based on whatever WB was selected when we took this "calibration" shot. Therefore, the system, presented with the shot, needs to be able to "back out" that WB correction to really determine the original average chrominance of the test "scene" (that is, in the ideal case, of the test card or the back of the ExpoDisc).
I'll press on as if my assumption in this regard were correct.
How can I determine those WB parameters from an existing file? A slightly different question: how can I determine the correction parameters for a certain WB setting in the camera (standard or custom)?
I assume that the parameters of the current WB setting are also captured in a CRW file. I say this because I understand that, by default, the RAW image conversion software applies to the derived image the same WB correction as would have been applied in the camera, for a JPG output, based on the current WB setting.
Will the various RAW conversion packages directly reveal to us those embedded WB correction parameters? If so, then to answer my second question, could I just take a shot of any old scene, present the CRW file to the RAW conversion software, and ask it to tell me what WB parameters t would apply if I didn't tell it otherwise?
Then, with respect to my first question, would there be any corresponding ploy that could be applied to a JPG file to determine the parameters of the WB correction that had been in effect when the shot was taken.
Thanks.
Best regards,
Doug
White Balance so easy, even our 5 year old can do it.- Melissa Strickland
I hope some of the forum gurus will answer. I shot primarily on JPEG, RAW just for those times when the dynamic range will be too much for the camera. Therefore a good answer to your question will add more tools to use in the digital darkroom.
White Balance so easy, even our 5 year old can do it.- Melissa Strickland
It must be a real mystery. I posted the same inquiry on the DPR 300D forum, and I haven't yet gotten a thing back, not even get the customary "well of course we don't know, why would that good-for-nothing bunch of ignorant thieves at Canon make it possible for us to see this vital-to-our-lives information, and by the way don't you think its a crime that the case of the 300D is made of plastic."
(Of course, I personally don't get many responses like that!)
Best regards,
Doug
White Balance so easy, even our 5 year old can do it.- Melissa Strickland
I've often wondered how/where they record the base RGB WB (white balance) data values in an image file used for CWB (custom white balance). Knowing Canon, if they consider the placement within the file or the method for storing this data proprietary, we may never know the answer unless someone reverse engineers the data. Unfortunately, from what I can tell, the SDK conversion routine just samples an x,y position within the CWB image to obtain WB values when converting images, and this might be the only value stored within RAW images files. I'd even say the same for jpg files if it weren't for the sequence used to assign a CWB value. It's done 'after' the capture used for CWB is written to a file, so it must be there somewhere.
I rarely shoot jpg, but now you've got my curiosity tweaked...
White Balance so easy, even our 5 year old can do it.- Melissa Strickland
A related - albeit slightly less complicated - thought:
Would other jpeg users consider a NO WB setting to be useful?
I have just got my first digital camera (a 300D) and have only spent an hour playing with RAW - at this stage I figure I'll learn more just by taking lots of (jpeg) photos.
So far I have not noticed any compression artifacts in my Large Fine jpeg files. In stark contrast, however, I have taken some shots in Auto WB that I would consider unrecoverable due to the horrible colour cast imparted.
Having read lots of posts it appears some RAW users do so primarily so that they can adjust WB later (yes, I know there are other advantages). Given that there is arguably a fair bit of extra work in a RAW workflow, wouldn't such users benefit from a basic firmware tweak that would allow for a jpeg to be recorded without any colour correction?
White Balance so easy, even our 5 year old can do it.- Melissa Strickland
I would assume the camera has to have *some* sort of WB setting. From what I understand (which admittedly isn't a whole lot LOL), WB is just a "reference" to what the camera will use as "neutral white." This shifts the color balance one way or the other, often referred to as "cool" or "warm" toned.
Without ANY WB setting, I would assume the camera would not know what white is, and the images would be unuseable.
FWIW, I find Auto WB to be what I would consider quite useable. It works much better than the first two digicams I owned (Kodak DC215 and Canon G1). Outdoors it is particularly reliable. Indoors, not so much. I have started using the coffee filter method inside (this weekend!) and am very pleased with the results. Outside I might stick with auto.
Interestingly enough, my Olympus C5000Z P&S does better at Auto WB indoors than my 300D... Ahwell, one area for Canon to improve.
White Balance so easy, even our 5 year old can do it.- Melissa Strickland
Just a quick summary; I cannot comment on JPEGs immediately as I haven't got the data to hand.
The way that the WB / CWB data is used in the D30/D60/10D/300D (guessing from intuition on the last one) is that the camera does AWB for *every* image, regardless of the WB setting. The full list of WB parameterers for all the settings (four params, 2 green 1 red 1 blue - to match the Bayer array) are stored with each CRW file in the CIFF data. I *think* this is in the JPEG data too, for exactly the reason of CWB settings.
What happens for CWB is that the AWB parameters are taken from the chosen image and used as the preset. Hey presto.
If my quick post confuses, send me mail and I will explain slower and when I have more time :-)
White Balance so easy, even our 5 year old can do it.- Melissa Strickland
Linear Mode
