What’s the best way to batch-remove dust and scratches from scanned slides?

Asked 1/4/2013

2 views

2 answers

0

I have a few hundred 30-year-old slides with visible dust and scratches. I’d like to automate as much cleanup as possible rather than retouching each image by hand. Are there batch-processing filters or tools for Windows or Linux that can remove most of this damage without destroying image detail? If software-only cleanup is limited, is there a better workflow for getting cleaner results from the original slides?

Originally by Photography Stack Exchange contributor. Source · Licensed CC BY-SA 4.0

Photography Stack Exchange contributor

13y ago

2 Answers

1

If you still have the slides, re-scan them with hardware dust removal. This is after you have cleaned them using canned air.

Rent a slide scanner which has Digital ICE. It's a method to remove dust by using an infrared light source and creating a "dust only" image which is then used to clean up the original image automatically. You can also just buy a scanner off of ebay and sell it for the same price when you are done. I've noticed that the late scanners by Nikon and Minolta are much better than the newer ones from Plustek or Reflecta.

Or send the slides to a scanning service like digmypics.

Originally by user15615. Source · Licensed CC BY-SA 4.0

user15615

13y ago

0

AI Answer

Generated from our catalog & community — verify before relying on it.

For large batches of dusty old slides, the best solution is usually not a software filter after scanning, but re-scanning with hardware dust removal.

If you still have the original slides, clean them first with canned air, then use a slide scanner that supports Digital ICE (or a similar infrared dust-removal system). It uses an infrared pass to detect dust and scratches and automatically removes much of it with far less detail loss than generic image filters. Renting a good film scanner or using a scanning service can be more effective than trying to batch-fix existing scans.

Software-only filters can help somewhat, but they’re a compromise. Dust/scratch removal generally requires selective masking and filters such as median filtering; simple blur filters like Gaussian blur will soften detail too much. Fully automatic batch processing on already-scanned images is likely to leave artifacts or reduce sharpness, especially on high-resolution scans.

So: if quality matters, re-scan with infrared dust removal. If that’s not possible, software cleanup may be acceptable for smaller display sizes or modest prints, but it won’t be as clean or as detailed.

UniqueBot

AI

13y ago

Your Answer