Based on this question, a question arose regarding the white halo that appears around the main object with the following conjectures:
Where A shows a white halo while B has no halo:
But, the same halos appear in the photographs of this link but only in objects with a light background, such as the sky, not in those that appear on a similar background:
Any answer?
It is not due to the lens.
For there to be anything discernable there must be contrast (color/brightness). And anything done to increase visibility (sharpness/clarity/dehaze/etc) is some form of increased contrast.
This particular example is characteristic of low level (large detail) sharpening... if it was a raw file it was done in pos. If it was recorded as a jpeg it could have been done at any point, but this degree of artifact is more typical of post editing. Understand that any image must undergo "editing" in order to be a visible image created from raw sensor data.
The primary clue here is that the contrast increase has created a light halo along dark edges, but also a dark halo (outline) along light edges.
You do not see the halo at point B because there is not a constant level of contrast along that edge... i.e. it is lost in the BG texture/details.
This halo is typical of sharpening methods that usually:
Lifted from an old Gimp documentation:
Using an unsharp mask to sharpen an image can seem rather weird. Here is the explanation:
Think of an image with a contrast in some place. The intensity curve of the pixels on a line going through this contrast will show an abrupt increase of intensity: like a stair if contrast is perfectly sharp, like an S if there is some blur.
Now, we have an image with some blur we want to sharpen (black curve). We apply some more blur: the intensity variation will be more gradual (green curve).
Let us subtract the blurredness intensity from the intensity of the image. We get the red curve, which is more abrupt : contrast and sharpness are increased. QED.
What the Gimp doc doesn't tell but is visible in the picture above is that the red line "overshoots". In other words, near the place where the transition happens, there is are small areas where things are darker/lighter than they should be. If the transition is along a line (marked edge of an object), these areas from a halo (usually the light halo is much more visible than the dark one).
Unsharp mask has first been used in silver photography. The photograph first creates a copy of the original negative by contact, on a film, placing a thin glass plate between both; that will produce a blurred copy because of light diffusion. Then he places both films, exactly corresponding, in a photo enlarger, to reproduce them on paper. The dark areas of the positive blurred film, opposed to the clear areas of the original negative will prevent light to go through and so will be subtracted from the light going through the original film.
