DUST & SCRATCHES
I may be the last one to have figured this out but it’s new to me so I’m posting it anyway. It’s really easy to get rid of the “speckly” noise I pointed out in the last couple posts using the Dust & Scratches filter in Photoshop. My testing has showed that it results in no loss of image quality and doesn’t smear details like “normal” noise reduction generally does. It also works considerably better than the Despeckle filter which gives you no control and visibly blurs details.
Here is a 300% zoom crop (it’s visible way below 300% but I just wanted to make it easy to see) before and after the Dust & Scratches filter was applied.
Dust & Scratches filter before and after, 23 minute exposure, ISO 400, Long exposure noise reduction ON, Settings: Amount 1, Threshold 10 (vary the threshold)
An important side note - this is a crop from a 23 minute exposure at ISO 400 taken in succession to three 15 minute exposures at ISO 400. Long exposure noise reduction was ON for all exposures and while the 15 minute exposures showed no speckling the 23 minute exposure did. There was definitely some threshold that was crossed during that extra eight minutes but I’m not sure if it was a result of the longer exposure or that the sensor was finally starting to heat up after more than an hour of exposures with little break in between.
The Dust & Scratches filter also works really well on the long exposure tests I posted earlier where the images without in-camera long exposure noise reduction showed speckling. Here is the 15 minute long exposure without noise reduction before and after the Dust & Scratches filter. I converted to black and white because it was easier to see the difference - in the color version there is still some color blotching left but it is far less noticeable than the speckling.
Dust & Scratches filter before and after, 15 minute exposure, Long exposure noise reduction OFF, Settings: Amount 1, Threshold 6 (vary the threshold)
Clearly the filter makes a significant difference here and until I see otherwise, for me, it’s justifiable to skip long exposure noise reduction with these results. Also for what it’s worth I don’t think this would work for astrophotography for obvious reasons.
