Spot removal

[Frodo]

I shot the attached image of a moonset over our beach yesterday.  As I wanted a long exposure I set the camera (Canon 5D) at the lowest ISO and the lens (24-105) at the minimum aperture (I did not have a neutral density filter and plan to get one).  So, not surprisingly I ended up with a lot of sensor dust on the image.  
I found that the spot correction tool in LR 3.2 is incredibly slow and seems to get slower the more spot corrections in an image (I had over 50, but that's another story).  Furthermore, I found that the corrections often generated artifacts because it is a cloning tool, even though I was careful where I shifted the source circle to.  
Anyway I soon got sick of this and removed the remaining dust spots (and some of the artifacts) with the spot healing tool in Photoshop Elements 7.  This was much, much quicker and produced a far superior result.
The one advantage of LR, is that the spot removals can be pasted in another image.  
But for me, I won't use LR to remove a large number of dust spots in the future.
Cheers

[Slobodan Blagojevic]

... it is a cloning tool...

And it can be switched to a spot healing tool just as well.

[Frodo]

And it can be switched to a spot healing tool just as well.

I have the tool set to "heal", but it still finds a section of the image nearby that it pastes into the selected spot.  In contrast Photoshop Elements the healing tool combines colour and texture from around the selected spot, rather than the clone tool, which pastes a section of the image from another (single) spot and so appears to operate like the spot correction in LR 3.2.  Or am I doing something wrong?

[Schewe]

Or am I doing something wrong?

You are misunderstanding the Lightroom spot removal tool's use and functionality...Lightroom (and Camera Raw) has a "Spot Healing" tool, not the full featured healing brush. As such the usability and how to use it is different that "healing" in Photoshop or Elements. It works well for spots but will not duplicate what the healing brush can do.

[pegelli]

Or am I doing something wrong?

As Jeff said don't think so, however by moving where the "base texture" is coming from I usually can do very well on removing dust w/o creating strange artifacts.

[stamper]

In Photoshop I believe that you don't sample using the spot removal tool, you just click on the spot you want to remove. It works best when there isn't, or there is little texture, surrounding it.

[PeterAit]

I don't understand why a low ISO and small aperture would result in sensor dust spots. Can you please explain?

[pegelli]

I don't understand why a low ISO and small aperture would result in sensor dust spots. Can you please explain?

The OP was using low ISO and a small aperture to achieve a long exposure.

Low ISO would normally not cause sensor dust to show more (unless noise at high ISO gets so bad it masks the dust spots  )
Small aperture increases the depth of field, so dust lying on the AA filter will become more in focus and visible versus a large aperture where it will more easily blend into the background.

[PeterAit]

That's not correct. Small aperture increases depth of focus in front of the lens, not behind it (a meaningless concept anyway).

[Jim Pascoe]

That's not correct. Small aperture increases depth of focus in front of the lens, not behind it (a meaningless concept anyway).

Actually (and I do stand to be corrected) I thought depth of field describes the range of focus in front of the lens and depth of focus the range of focus at the film or sensor plane.  Surely it is not a meaningless concept because if the depth of focus is very small the sensor (or film) needs to be positioned very accurately to give a sharp image.  Probably less of a problem with a sensor than a piece of film!

Jim

[pegelli]

That's not correct. Small aperture increases depth of focus in front of the lens, not behind it (a meaningless concept anyway).

Well the principle of the smaller pupil creating smaller circles of confusion is the same in front and behind the lens, so the effect on the dust spots lying a small distance of the focal plane is similar to an increased dof. I'm not going to quibble the right optics law terms, but I hope you get the picture  

[PeterAit]

No, that's not how it works. The image is coming in one direction, from in front (the scene) to in back (the sensor). Dust spots occur because the dust particles block the light from reaching the sensor - the spot is a shadow of the particle, not an image of the particle.

It is like projecting images on a screen. If you hold your hand in front of the screen, you get a shadow of the hand on the screen, not an image. It does not matter whether the image you are projecting is in focus or out of focus, the hand will still block light from reaching the screen.

Best,

[pegelli]

Peter, if you know the answer why did you ask the question?

If the pupil is smaller the beam is "sharper" hence the image or shadow sharper. I agree the optical science buffs might call it different, but for me as a layman it's the same principle as why dof increases with a smaller pupil.
Take a large softbox and a shadow of a small object is hardly noticable on a backdrop, if you take a point light source you get a very distict and shadow of the exactly the same object at the same distance.

If you don't believe dust is more visible at smaller apertures just do a test and post the results here, then we can discuss further.

[PeterAit]

Hmm, I think I see your point. The dust is not directly on the sensor but on the filter that is a small distance in front of it. Thus, a diffuse light source casts a less distinct shadow than does a sharper light source.

I would do some tests but I am blessedly free of dust in all my cameras. Maybe I don't get out enought!

[eliedinur]

I would do some tests but I am blessedly free of dust in all my cameras. Maybe I don't get out enought!

Have you examined them with f22 or smaller opening and shooting a blank surface?

[Frodo]

I would do some tests but I am blessedly free of dust in all my cameras.

Dust is normally not a problem for me as I tend to shoot wide open with long lenses and at middle apertures for wide lenses.  However, I needed a long exposure to blur the waves for this image, hence the low ISO and f22.  I had to remove about 50! spots from the image.  Most were tiny, but were quite visible given the lack of detailed texture over most of the image.  In a normal landscape, probably only a few of these would have been visible.

As elied noted, you may be surprised by shooting a test as described.

Cheers

[feppe]

Dust is normally not a problem for me as I tend to shoot wide open with long lenses and at middle apertures for wide lenses.  However, I needed a long exposure to blur the waves for this image, hence the low ISO and f22.  I had to remove about 50! spots from the image.  Most were tiny, but were quite visible given the lack of detailed texture over most of the image.  In a normal landscape, probably only a few of these would have been visible.

As elied noted, you may be surprised by shooting a test as described.

Cheers

For future reference, it's usually advisable to use a good ND filter instead of going to f22 - diffraction will enter much earlier than f22 to degrade image quality more than the filter.

[Frodo]

Feppe,

Quite correct.  I didn't have one with the right diameter at the time.