How can I select bright pixels by threshold in GIMP 2.9 for astrophotography?
Asked 7/10/2017
5 views
2 answers
0
I'm editing a 16-bit astrophotograph in GIMP 2.9 and want to isolate saturated stars or other very bright pixels above a chosen threshold. The usual Select by Color approach isn't suitable for this. Is there a way in GIMP 2.9 to create a selection or mask based on pixel brightness/value, such as thresholding everything above a certain level?
Originally by Photography Stack Exchange contributor. Source · Licensed CC BY-SA 4.0
Photography Stack Exchange contributor
9y ago
2 Answers
1
The way to do this in gimp 2.9, as suggested by a post on the gimp users mailing list here is to threshold on a grayscale duplicate (or channel depending on how you want to do it); then that can be turned into a selection or mask.
Originally by user65606. Source · Licensed CC BY-SA 4.0
user65606
9y ago
0
Generated from our catalog & community — verify before relying on it.
Yes. In GIMP 2.9, a practical way is to threshold a grayscale copy rather than use Select by Color.
Typical workflow:
- Duplicate the image, or work from a channel if that better represents the brightness you want.
- Convert that duplicate to grayscale if needed.
- Apply a Threshold operation to isolate the brightest pixels/stars above your chosen cutoff.
- Turn the resulting black-and-white result into a selection or use it as a layer mask.
This works well for astrophotography because you’re selecting by luminance/value instead of by color similarity. If you only need the brightest saturated stars, thresholding a grayscale version is usually the simplest route.
Recommended products
UniqueBot
AI9y ago
Your Answer
Related Questions
Do any image sensors measure per-pixel time-to-saturation to increase dynamic range?
How can I cleanly mask a subject and thin rope in Affinity Photo?
Can Canon DPP automatically set exposure so a clicked spot does not exceed RGB 250?
How can I maximize contrast for a grayscale machine-vision camera tracking reflective strips?
How do I map pixel coordinates in a scanned star photo to RA/Dec?