V1.02
Thom Hogan’s Complete Guide to the Nikon D300
Page 273
All the same things I said about the luminance histogram
apply to the individual channel histograms. Spikes at the right
edge are blowouts in the highlights (in the case of an
individual channel that would be called a “channel blowout,”
as in “I blew out the Red channel”). Spikes at the left edge
mean shadow detail is lost in that channel.
So why are channel histograms important? Remember those
nuclear colors I mentioned before? Well, channel histograms
would be the one tool on the camera that might alert you to
the fact that you’ve got one. But even in some situations
where you might not be expecting it, the channel histograms
can save you from an exposure error.
The classic example for my type of photography is the red
rock country of the US Southwest (Northern Arizona and
Southern Utah, for example). If you take a picture of a
landscape feature that’s in bright sun, all that red in the rock
has a tendency to push the red channel up, often enough to
blow the channel out completely. This has an impact after the
fact: when you go to post process that picture and perhaps try
to alter the white balance a bit, the blown red channel will
prohibit you from many manipulations you might want to do,
and the tonal ramp in the areas that have been blown out may
be compromised. Unfortunately, the luminance histogram
alone may lie to you in these situations, saying that the
exposure is okay.
Red rock isn’t the only thing you have to watch out for,
though, which is why you just have to pay attention to the
channel histograms. Any red or blue channel blowout means
you need to reduce exposure. Green channel blowouts are a
little less likely to occur and less problematic
86
, but if you
have a large expanse of green in bright light (like a lawn on a
golf course) you may still need to bring the tonal value down.
86
That’s because most color manipulations you’d make after the fact impact the blue
and red channels more than the green, which is in the middle of the spectrum
between the two.
