Flare is no aberration, but it may impair the image more than the residual aberrations in a photographic lens. Manifestations of flare are diverse and range from colored patches, ghost images, haloes, a haze over the entire image, and any mixture or variation. Flare is due to light that hits the film at the wrong point or that was not supposed to reach the film in the first place. The effect is most obvious in dark areas of the image. The causes of flare are as varied as its manifestations. Reflections at air/glass interfaces, at the lens barrel, at the diaphragm blades, at the film, scattering at impurities within the glass, or even camera light leakage. The advent of (multi)coatings greatly reduced the magnitude of reflections at the air/glass interfaces of lens elements, but the performance of a coating is not constant over the spectrum and depends on the angle of incidence. Residual reflections are of order 0.5 to 1%, compared to ~4% for an uncoated surface.
A common provocation for a photographic system is the presence of the sun within or outside the lens angle of view. Since the sun is so much brighter than the photographed scene, only a small fraction of the direct sunlight needs to be scattered to be noticed on top of the actual image. Part of the responsibility to minimize flare rests with the lens designer. Proper lens coatings reduce the reflections at air/glass interfaces and the various mechanical parts should be efficiently blackened. The photographer, for his part, should use a proper lens hood to eliminate all light that is not required to record the actual scene. He should avoid dirt or fingerprints on optical parts and be careful with filters.
Figure 1 illustrates the occurrence of lens flare and the usefulness of a proper lens
hood. There is a regular light ray that belongs to the photographed scene, this is the
blue ray that arrives within the lens angle of view indicated by the brighter gray. This
ray is normally refracted and contributes to the image formation. (Note however that I did
not attempt genuine ray tracing, the light paths in figure 1 are merely illustrative.)
There is also a slanting ray of sunlight that impinges upon the front element when a lens
hood is absent. A small part of the light is reflected from the rear surface of this
element (red arrow). A fraction of this light is reflected again from the front surface,
directed to the film (dashed red line). This is flare. Figure 1 shows a short lens hood
(the built-in hood of that lens) which does not block the slanting ray. The wider
and longer hood does intercept this ray and does a better job at
flare protection as it simply blocks more undesired light than the shorter hood. A more
detailed discussion on the size and shape of lens hoods is found on the
lens hood page.
To be sure, figure 1 exemplifies just one of many causes of flare. Numerous reflections are possible, at all optical and mechanical boundaries. The regular ray in figure 1 may also introduce flare by reflections off the lens elements, which cannot be prevented with a lens hood.
When reflections occur in the rear group of elements, a flare effect known as aperture ghosting results. A series of images are formed near the film plane. On the film, the consequence is a streak of patches of various sizes and colors, the shape of which resembles the diaphragm opening. Figure 2 exemplifies this kind of flare. The sun struck the lens through a roof window.
A photographer should be alert with veiling glare on the lurk. The veil is not always noticed in the viewfinder, but yet relegates a brilliant photographic lens to an inferior piece of glass. It usually originates from reflections in the front group of elements (cf. figure 1) which, by the time they reach the film, are spread out over a large area. The result is a whitish haze that covers the entire image, or large parts of it. The contrast of the image is reduced in parallel with the color saturation, an effect which is well known with sunlit television or computer screens. Veiling glare appears to various degrees, but even the slightest haze deprives the lens from its brilliance. The left image in figure 3 is seriously affected by veiling glare. It was taken on a blue sky day with the sun freely illuminating the front element of a 2.8/135. The built-in hood was inadequate to block the direct sunlight, so I used my hand as an effective shield (right image). The presence of the right image illustrates to better advantage the damage done to the left one.
Arthur Hood captured a UFO flying in over the pier of Brighton (figure 4). Did he? Well, no. The object in the air is in actual fact a mirror image of the building on the pier. Its windows are so bright that ghosts pop up in the dark sky.
In Arthur's case the flare is due to the protective filter that he used on his Carl Zeiss 4/80–200. The flare is a mirror ghost of the original object, with the image center serving as the point of symmetry. All dimensions are perfectly preserved, which suggests that reflections at planar surfaces are responsible. I worked out an explanation in figure 5. Black arrows indicate the light rays of a distant source that form a regular image point on the film. Values for the reflectance of undeveloped photographic film vary from 15% to 40% [1,2], which makes the film a much stronger reflector than any optical component in the lens. So, a significant percentage of the light is reflected off the film, partly specular and partly diffuse. To follow a few paths of the reflected light, it is convenient to consider the paths which are already drawn for the incident light. Thus, the blue arrows indicate light reflected from the film. This light encounters the filter, which specularly reflects a small fraction (red arrows). The red rays are parallel and consequently focused onto a point on the film. The virtual source of the mirror point is traced by the dashed black lines. Note that the blue rays reflected by the film seem odd from the viewpoint of specular reflection; they merely illustrate the fact that all light rays that orginate from a single point on the film, and which are collected by the lens, emerge parallel at the filter.
I experimentally verified that the
presence of a filter is indeed an invitation for this kind of flare.
Mirror ghosts do not necessarily occur because a filter is optically
inferior, but rather because it is a planar component.
It already appears from figure 5 that not all light reflected off the film makes it back to the mirror point. The presence of an aperture stop further reduces the number of rays allowed to return to the film. People who want to try their luck with UFOs may improve their chances by using a tele lens at a large aperture. I suppose that, instead of a filter, a lens element with a flat face could also give rise to mirror ghosts.
Under normal circumstances a protective filter does no visual harm to the
image, but it has become clear that it falls short under critical circumstances.
The importance of a lens hood cannot be emphasized enough. Zoom lenses are
provided with a lens hood that offers reasonable protection at the wide end, but
which is inadequate at the tele end. Likewise, the built-in hoods of tele
lenses are invariably too short. Wider and longer is
better. Often manufacturers do not even sell proper lens hoods for their tele
lenses, because these would be unwieldy and make the system susceptible to the
wind. In all these cases the furnished hood is better than nothing, but to get
the best out of the lenses it is recommendable to use additional shielding. A
flexible yet Spartan solution is a compendium hood, which allows for precise
adjustments to suit a wide range of focal lengths.
Personally, I consider veiling glare the most detrimental form of flare, whereas a nice streak of colored aperture ghosts could be used to good artistic effect. And wouldn't life be boring without UFOs? Opinions differ however, as other authors find veiling glare the least disturbing manifestation of flare [3, 4].
© Paul van Walree 2001–2014
|||SPSE handbook of photographic science and engineering, edited by Woodlief Thomas Jr., John Whiley & sons, p. 204 (1973).|
|||Sidney F. Ray, Applied photographic optics, 3rd ed., Focal Press, p. 139 (2002).|
|||Arthur Cox, Optics: the technique of definition, 6th ed., Focal Press (1946).|
|||Focus Elsevier foto- en filmencyclopedie, 4e editie, Amsterdam/Brussel (1981).|