Astrophotography diagnosis of M31: Oversaturation, residual vignetting and 2 others
Processed
The Doc examined this image of M31 (processed). Estimated overall technical quality: 7/10. 4 defects found: Oversaturation (severity 3/5), Residual vignetting (severity 2/5), Clipped stars (severity 2/5), Over-denoising (severity 2/5).
&w=1920&q=75)
Info
- Cible
- M31
- Position
- 0h42m43s · +41°16'12"
No Moon at all, so no lunar factor to weigh: this is the best possible window for a broadband target like M31, whose outer extensions are faint and highly sensitive to any stray luminance. The measured background is consistent with a calm sky, with no significant oriented gradient (plane amplitude only 12%, R² 0.18, so no directional light-pollution signature). The only real background modulation is radial, hence instrumental rather than atmospheric. Under these conditions the limiting factor is not the sky but integration time and a gentler processing hand: use these moonless nights to accumulate signal on M31's outer extensions instead of compensating with smoothing.
Setup
- Type d'image
- Traitée
- Phase de lune
- 0
- FOV
- 3.00°
The 3.00° field is well sized for M31: the galaxy spans about two thirds of the width and M110 fits inside the frame, leaving room for the faint extensions. The framing does place M31 rather high with M110 alone in a large empty area at the bottom, so shifting the center down by half a field would balance the composition better and pick up M32. Optically, measured FWHM between 2.5 and 3.1 px with elongation peaking at 1.07 in the periphery indicate a well tuned optical train: correct backfocus, flat sensor, matched corrector. On a monochrome sensor, keep this configuration, there is nothing to fix on the hardware side.
The diagnosis in detail
The verdict is clear: hardware and acquisition are good, everything degrading this image comes from post-processing. The PSF panel is remarkably homogeneous, with center elongation at 1.02 (exactly the field floor), a corner/center FWHM ratio of 0.866 (corners are even tighter than the center, where a bright star inflates the measurement) and horizontal and vertical asymmetries below 0.035. No signature of tilt, backfocus, coma or tracking drift: the 57° PA dispersion confirms round stars whose angle no longer carries meaning.
The common thread among the retained defects is an over-energetic stretch. M31's bulge has gone to a neutral white plateau, while the DSS shows a structured, colored nucleus in the same place, and the bright stars followed the same path with cores lacking any chromatic gradient. At the bottom of the range the opposite move was applied: the background was pressed against the black point (measured margin of only 1.5) and then smoothed, which erases grain but also the faint extensions of M31's halo, the very thing that makes this target rewarding.
That leaves the 24% radial falloff, rotationally symmetric, so a vignetting not fully settled by the flats rather than a sky gradient. It is the easiest point to fix and is best dealt with before any stretch, otherwise it gets baked into the final image.
Priority actions
- Reprocess from the linear master with a highlight-protecting stretch, to restore structure and hue in M31's bulge
- Correct the residual 24% vignetting before the stretch, with a re-shot matching flat set or a multiplicative background correction
- Raise the black point to leave about 5 to 10 levels of margin for the sky background and recover the faint halo extensions
- Greatly reduce denoising and mask it onto the low-signal areas to keep a natural grain
- Add a short-exposure set to rebuild the cores of the bright stars
