Light Problems

[Pauky]

See the purple reflected light at the bottom? What is it and what causes it? How can I correct it? Thanks for comments.

http://members.cox.net/ewhistler/issue.jpg

 

[Pauky]

bump

 

[smyth]

chromatic abberation possibly, but i doubt it. something in the water maybe?

 

[pacman]

Is that oil on the surface of the shore?

 

[j_mcquillen]

It looks more like chromatic aberration or something similar to me... perhaps your camera has been unable to cope with particularly bright highlights?

The colour doesn't look like oil or a reflection or anything else natural to me - have you noticed this before in any of your pics?

 

[Alex_B]

It looks more like chromatic aberration

not at all ... CA is a lens effect which has nothing to do with highlights plus it should be stronger the further away you get from the centre of the image.... but here it is only at the bottom.

CA results from light being refracted at different angles depending on the wavelength (colour).

.. perhaps your camera has been unable to cope with particularly bright highlights?

much more likely

 

[ksmattfish]

It's called "purple fringing". Just google the term, and you'll find lots of articles about it, why they think it happens, and how to deal with it.

I have noticed purple fringing with some of the cheap sunglasses I wear, so I wonder if it may have something to do with in-camera filters or cheaper multi-coatings.

 

[Hertz van Rental]

It's technical name is lateral chromatic aberration.

 

[Alex_B]

It's technical name is lateral chromatic aberration.

Is it? did not know that.

If what is shown here is a sensor effect though, then it would not be an aberration in an optical sense, would it? Or do you suggest this is lens related?

just asking

 

[Alex_B]

Again, to me this cannot be lateral CA,

1. because it sets in suddenly, and does not increase gradually with distance from the optical axis (which is hopefully approximately in the centre of the image ).

2. because it is extremely asymmetric.


It could still be axial CA, which causes fringes to appear all around dark objects on bright blown out background, independently of the distance to the optical axis of the system, or a sensor effect. since in this image the effect seems to be strongly depending on the brightness, I vote for the latter.

 

[Hertz van Rental]

I didn't say this was caused by lateral chromatic aberration, only that it was the technical term for purple fringing.
Axial CA would cause purple fringing on both sides.

I'm not sure about this, other than it is an effect caused by the camera and/or lens.
CA usually produces a colour more bluish - this is very purple. It could be CA of either type working on the specular highlights - red on one side and blue on the other and on such small points might give the appearance of purple - but it's very localised and doesn't appear to be on other portions. CA normally makes itself apparent where you have a boundary between high contrast and low contrast areas. There is no sign of it underneath the gull. CA normally gets progressively worse as you move away from center - this has a sharp boundary. I would therefore be tempted to rule out the lens as the cause.
Similarly I would rule out anything in the scene itself (like oil) as you would expect blue and yellow in a slick.
This leaves us with the camera. I presume it is a digital as I have never seen anything quite like this with film. I can only imagine, looking at what is turning purple, that it is an effect caused by localised gross over-exposure (and I'm just airing theories here) producing interference - or it could be a processing problem (as in software processing).
But who knows?
This is one of the joys of Photography. It never does quite what you expect it to

 

[Alex_B]

I didn't say this was caused by lateral chromatic aberration, only that it was the technical term for purple fringing.
Axial CA would cause purple fringing on both sides.

There are two things one could mean by both sides:

"both sides of the whole field of view", or "both sides of a small structure, e.g. a branch against the sky, a small isolated light source, and so on" This latter thing refers to something local.

lateral CA is one sided locally, but occurs on both sides of the whole field of view (actually all four sides)

axial CA is both-sided locally, but with respect to the whole image it will be weaker towards the edges of the frame, but not much, ... mainly it is most prominent where the strongest contrast is.

CA usually produces a colour more bluish - this is very purple.

For an uncorrected lens, the blue-ish CA is the strongest, but with most decent lenses today that part of the spectrum is well compensated by design (achromatic or what it is called, ED glass with Nikon and L glass with Canon).

but not many lenses compensate well for other parts of the spectrum. Those which do (apochromatic I think is the term here), cannot be afforded by most of us

This strong purple fringing as it appears in the image presented, independend on the distance from the optical axis and very pronounced at stron contrast edges with blown out highlights, I only know from early p&s cameras with small sensors.

BTW, not trying to act as a smart*ss here, just trying to find convergence and enlightment

 

[Garbz]

Easy way to rule out the camera and the optics. Take a picture that produces the same result, then turn the camera upside down and take it again.

I'm perplexed that it is so severe to the bottom of the frame yet the contrasty transitions at the top have NO fringing at all.

 

[Alex_B]

I'm perplexed that it is so severe to the bottom of the frame yet the contrasty transitions at the top have NO fringing at all.

yes, that is the main problem with this one ... could be:

a) this is a crop and not the full frame, however, still the transition from no purple to purple is very abrupt

b) that area is really so much brighter, reflecting
the sun exactly into the camera ...

 

[Hertz van Rental]

yes, that is the main problem with this one ... could be:

a) this is a crop and not the full frame, however, still the transition from no purple to purple is very abrupt

b) that area is really so much brighter, reflecting
the sun exactly into the camera ...

It's the abrupt transition that makes me think it's a sensor issue.
And there appears to be minimal flare and related effects.

Of course, there is the outside possiblity that it has been done on purpose with PS just to get us going :lmao: