Flower color accuracy problems

I've been struggling with my camera's tendency to get flower colours
wrong where blue purple and magenta are concerned (Sony A550). Not in
any consistent way. Sometimes blues come out purple, sometimes purples
come out blue. It's partly that auto white balance gets it wrong when
the flower bloom fills the image. Using a gray card to set white
balance improves things. It's partly intense colours getting
overexposed -- reducing exposure helps.

So by way of experiment I've been photographing flowers in the garden,
processing and printing them, and then taking the print out to the
garden to compare it with the flower. I was stunned to discover that
my latest careful methodology had produced a blue print of a very
obviously pink Morning Glory bloom.

But wait a minute! I was sure I remembered the flower being blue!
Could it be changes in UV radiation?

It turns out that some Morning Glory flowers change from blue to pink
during the course of a day. And back again. Hm...

--
Chris Malcolm

Chris Malcolm wrote:

But wait a minute! I was sure I remembered the flower being blue!
Could it be changes in UV radiation?

That would seem very likely to me. One way to check would be to use
an effective UV filter. If you have a Kodak 2A gel or a Tiffen 2A
filter one would almost certainly fix the problem if it is the UV
content that is causing the problem. (These filters have a slight
yellow colour to them, but are known to be completely effective
against UV. The fact that you can not tell a good UV filter from
an ineffective one by looking can be a problem because some things
sold as UV filters nowadays are sold for "protection" and the makers
do not expect the customer to care whether they filter out UV or not.)

Peter.
--

On 27/08/2012 19:30, Chris Malcolm wrote:

I've been struggling with my camera's tendency to get flower colours
wrong where blue purple and magenta are concerned (Sony A550). Not in
any consistent way. Sometimes blues come out purple, sometimes purples
come out blue. It's partly that auto white balance gets it wrong when
the flower bloom fills the image. Using a gray card to set white
balance improves things. It's partly intense colours getting
overexposed -- reducing exposure helps.

So by way of experiment I've been photographing flowers in the garden,
processing and printing them, and then taking the print out to the
garden to compare it with the flower. I was stunned to discover that
my latest careful methodology had produced a blue print of a very
obviously pink Morning Glory bloom.

But wait a minute! I was sure I remembered the flower being blue!
Could it be changes in UV radiation?

It turns out that some Morning Glory flowers change from blue to pink
during the course of a day. And back again. Hm...

It is a side effect of getting flesh tones exactly right. They
deliberately bury all the residual errors in the line of purples.

http://en.wikipedia.org/wiki/Line_of_purples

You have to be unlucky about the choice of plant but some do really mess
up slide film - much less so on digital CCD imaging.

--
Regards,
Martin Brown

On 8/27/2012 2:30 PM, Chris Malcolm wrote:
I've been struggling with my camera's tendency to get flower colours
wrong where blue purple and magenta are concerned (Sony A550). Not in
any consistent way. Sometimes blues come out purple, sometimes purples
come out blue. It's partly that auto white balance gets it wrong when
the flower bloom fills the image. Using a gray card to set white
balance improves things. It's partly intense colours getting
overexposed -- reducing exposure helps.

So by way of experiment I've been photographing flowers in the garden,
processing and printing them, and then taking the print out to the
garden to compare it with the flower. I was stunned to discover that
my latest careful methodology had produced a blue print of a very
obviously pink Morning Glory bloom.

But wait a minute! I was sure I remembered the flower being blue!
Could it be changes in UV radiation?

It turns out that some Morning Glory flowers change from blue to pink
during the course of a day. And back again. Hm...

But some flowers are notorious for photographing pink because of UV
while appearing blue to the eye; wild gentians of Switzerland are a
spectacular example.

--
Jim Silverton (Potomac, MD)

Extraneous "not" in Reply To.

On Mon, 27 Aug 2012 19:36:21 +0000 (UTC), Peter Irwin wrote:
: Chris Malcolm wrote:
:
: But wait a minute! I was sure I remembered the flower being blue!
: Could it be changes in UV radiation?
:
: That would seem very likely to me. One way to check would be to use
: an effective UV filter. If you have a Kodak 2A gel or a Tiffen 2A
: filter one would almost certainly fix the problem if it is the UV
: content that is causing the problem. (These filters have a slight
: yellow colour to them, but are known to be completely effective
: against UV. The fact that you can not tell a good UV filter from
: an ineffective one by looking can be a problem because some things
: sold as UV filters nowadays are sold for "protection" and the makers
: do not expect the customer to care whether they filter out UV or not.)

I'd expect flaws in the glass to be a greater problem with cheap filters than
failure to block ultraviolet. Most cheap glass doesn't pass ultraviolet well.
Two examples: Hospitals and nursing homes buy (or at least did the last I
knew) special UV-passing glass for their sunrooms. And photochromic sunglasses
don't work well in automobiles because the windows filter out most of the UV,
and it's UV that makes the glasses darken.

Bob