5
\$\begingroup\$

Background: This question is largely inspired by one of the answers in: What limits the size of digital imaging sensors?

So the answer mentions uber-large image sensors that are made by "quilting" or "tiling" smaller CMOS sensors in a single silicon. Now, I've seen quite a few astronomy images that uses this kind of humongous sensor.

One of the example image is this: enter image description here

Question: How can you eliminate the horizontal and vertical strips of missing photo? Is this kind of thing even possible or the astrophotography photos are heavily photo shopped?

\$\endgroup\$
3
  • \$\begingroup\$ What is the source of that image?? \$\endgroup\$ Commented Jun 29, 2015 at 15:15
  • \$\begingroup\$ @DigitalLightcraft pan-starrs.ifa.hawaii.edu/public/design-features/cameras.html \$\endgroup\$ Commented Jun 30, 2015 at 2:23
  • 1
    \$\begingroup\$ If you move on to the "Data Handling" page in the above link, it states: "The images from the four cameras are then compared with each other to remove spurious effects such as cosmic rays, gaps between CCD cells, and bad pixels, and a composite image is calculated based on the four subimages and corrected for measured variations in atmospheric seeing." \$\endgroup\$ Commented Jun 30, 2015 at 8:34

1 Answer 1

1
\$\begingroup\$

How to do it will depend on the width of the lines. Narrow ones, for "pretty picture" purposes can be healed by using basic nearest neighbour calculations. (Obviously this can't be used for scientific analysis of the raw data!)

Phase One, in some of their earlier sensors, had "bad pixel columns", which appear as a thin black line in the resultant images if not corrected. They "heal" these in software using weighted nearest neighbour calculations.

\$\endgroup\$

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

Not the answer you're looking for? Browse other questions tagged or ask your own question.