Tabular read of stats of multi images at once?

[gboyle]

I'd like to be able to eyeball quickly the stats on large numbers of images (mean pixel values and standard deviation) to assess image quality and decide if any images should be discarded.  Both the file explorer tab and the stats process seem to handle one image at a time when something more like the blink process with tabular info attached would be helpful.
Am I missing a trick here?
Thanks

[Juan Conejero]

Hi,

The Blink tool has a series analysis routine that provides the functionality that you want. For image grading, the SubframeSelector script is the most comprehensive and versatile tool available.

On a side note, you have much better estimators than the mean and the standard deviation in PixInsight. For location (central value) and scale (variability) we provide a number of robust estimators that give much more accurate and useful information on images. For example, Blink's series analysis provides median and median absolute deviation (MAD) estimates.

[gboyle]

Excellent.  I look forward to learning more about these powerful routines and stats.
Thanks
Gavin

[gboyle]

Juan thanks for your post.  is there somewhere i can read up on how to assess the quality of subframes (or indeed master frames)  using the stats available in the blink tool?  i'd like to be able to do a sense check on the masters generated and to spot any poor subframes.  i attach a report generated in blink for a few master darks we have taken - i note that the median adu count is the almost the same for the whole range of 60 secs to 900 secs - and the MAD is exactly the same for several of the dark masters- what should i take that as meaning?
i see that the subframeselector tool has good documentation but it does not seem suited to "eyeballing" calibration frames with no stars in them.
i'm happy to invest time getting to understand these calibration issues as am  hoping to do some science work before long!
thanks again

[gboyle]

and maybe life just doesn't work this way, but it would be great if  a version of the visuals from the blink process functionality could be embedded in the subframeselector process!!
 

[Phil Leigh]

I do find it intensely annoying that you can't have multiple instances of the Image Statistics process open in a single workspace...