What are barrel and pincushion distortion, and how are they corrected?

Asked 4/18/2011

20 views

2 answers

0

I understand barrel distortion as straight lines near the edge of the frame curving outward, which is common with wide-angle lenses. Is pincushion distortion the opposite, where lines curve inward? What types of lenses or zoom settings tend to show pincushion distortion?

I also know some cameras and software can correct these distortions. How is that correction typically done in practice, and is it usually applied before or after demosaicing RAW Bayer data?

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

Photography Stack Exchange contributor

15y ago

2 Answers

7

  1. Your interpretation of distortion is correct. For an example of a lens that exhibits barrel distortion at the wide angle and pin cushion distortion at the telephoto end, have a look at the Canon 18-135mm
  2. Tools like PTLens and Hugin perform the corrections after the demosaicing of the Bayer data. I'm not sure why this would matter?

If you're interested in understanding how this correction is done, the equations are presented here.

Originally by user2528. Source · Licensed CC BY-SA 4.0

user2528

15y ago

0

AI Answer

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

Yes. Barrel distortion makes straight lines bow outward toward the frame edges; pincushion distortion is the opposite, with lines bending inward. Barrel distortion is commonly associated with wide-angle settings, while pincushion distortion often appears at the telephoto end of some zoom lenses. A single zoom can show both depending on focal length—for example, some lenses show barrel distortion wide and pincushion distortion when zoomed in.

Correction is typically done by remapping the image geometry using a lens distortion model, so pixels are shifted to straighten lines. Software such as PTLens and Hugin applies this correction after demosaicing the Bayer data rather than directly in the raw Bayer mosaic. In general, the process is a geometric transform based on known lens characteristics, followed by interpolation to create the corrected image.

UniqueBot

AI

15y ago

Your Answer