Doug Kerr
Well-known member
I have recently completed a long-stalled technical article, "White Balance Color Correction in Digital Photography", and made it available on my technical information site, The Pumpkin. It is available here:
http://doug.kerr.home.att.net/pumpkin/index.htm#WhiteBalance
The article is comprehensive and detailed, but not mathematical in its outlook.
It begins with a review of the colorimetric and perceptual issues that give rise to te need for white balance color correction. Next, the basic concept is described, based on the notion of theoretically ideal color correction, although care is taken to point out that attaining this is not always our objective. This discussion illuminates the dependence of the process, in each case, on knowledge of the chromaticity of the incident illumination upon the subject.
Then, the two basic situations of white balance color correction are described, in-camera color correction (effective on an image output, such as JPEG) and color correction done during "external" development of the image from a raw data output from the camera.
We then discuss two approaches to chromaticity measurement with the camera: using a neutral target ("gray card") and using a measurement diffuser.
Some desirable properties of measurement diffusers are the explained.
We also discuss the two possible techniques of measurement with a diffuser: measurement at the subject location, and measurement at the place the camera will be for the actual "shot". The rather shaky "conceptual" premise for the latter technique is discussed.
Some mention is made of a commercial diffuser whose specifications "intentionally" depart from those often thought desirable, said by the manufacturer to give it a unique advantage in measurement "from the camera position".
Finally, in an appendix, we discuss the non-trivial issue of which way should a reflective target ("gray card") face, and the comparable issue of the orientation of a diffuser. In the course of this, we explains the role of "the cosine of the angle of incidence", which often appears in theoretical discussions of various aspects of this topic.
http://doug.kerr.home.att.net/pumpkin/index.htm#WhiteBalance
The article is comprehensive and detailed, but not mathematical in its outlook.
It begins with a review of the colorimetric and perceptual issues that give rise to te need for white balance color correction. Next, the basic concept is described, based on the notion of theoretically ideal color correction, although care is taken to point out that attaining this is not always our objective. This discussion illuminates the dependence of the process, in each case, on knowledge of the chromaticity of the incident illumination upon the subject.
Then, the two basic situations of white balance color correction are described, in-camera color correction (effective on an image output, such as JPEG) and color correction done during "external" development of the image from a raw data output from the camera.
We then discuss two approaches to chromaticity measurement with the camera: using a neutral target ("gray card") and using a measurement diffuser.
Some desirable properties of measurement diffusers are the explained.
We also discuss the two possible techniques of measurement with a diffuser: measurement at the subject location, and measurement at the place the camera will be for the actual "shot". The rather shaky "conceptual" premise for the latter technique is discussed.
Some mention is made of a commercial diffuser whose specifications "intentionally" depart from those often thought desirable, said by the manufacturer to give it a unique advantage in measurement "from the camera position".
Finally, in an appendix, we discuss the non-trivial issue of which way should a reflective target ("gray card") face, and the comparable issue of the orientation of a diffuser. In the course of this, we explains the role of "the cosine of the angle of incidence", which often appears in theoretical discussions of various aspects of this topic.