Strange error code in image integration.

[ballyhoo]

trying to run image integration, after having completed CC, DEB, SA, and the following error comes up:

*** Error: F:/PI/7000/sa/7000-3_002-3_c_cc_d_r.xisf: Zero or insignificant signal detected (empty image?)
<* failed *>


There certainly is an image(s) to integrate however. What in the world is going on here?  anyone have a clue? thank you so much!

The image I have on the screen (not detected) I applied STF for the purpose of being able to see the signal.
 

Here is a link to the star aligned file -- one of 15 or 16 files trying to integrate but the one the error message indicates to.

https://www.dropbox.com/s/miybw6m9qom9a47/7000-3_002-3_c_cc_d_r.xisf?dl=0

[bulrichl]

The data of this file are severely clipped, supposedly the calibration was done improperly. This is the statistics:

Code: [Select]

_7000_3_002_3_c_cc_d_r
            R            G            B
count (%)   51.57688     55.76581     39.35488      <==
count (px)  6031590      6521459      4602305
mean        241.380      147.820      152.754
median      0.000        0.000        0.000         <==
variance    4227188.736  2588166.957  3189147.884
stdDev      2056.013     1608.778     1785.819
avgDev      302.522      185.263      191.447
MAD         0.000        0.000        0.000
minimum     0.000        0.000        0.000
maximum     65534.996    65534.348    65534.996

Count should be near 100 %, median must not be zero. Therefore the error message is qualified.

Bernd

[dave_galera]

Zoom into the image....it looks very strange indeed.....

[ballyhoo]

Thank you for responding. I am not sure where I am going wrong. here is the next file in the sequence. Without the previous sub, the integration is okay.

https://www.dropbox.com/s/hafdsml7job846f/7000-3_004-3_c_cc_d_r.xisf?dl=0


and this is my settings for image calibration ---- but maybe the problem is flat calibration, where I try and subtract flats. Not sure

I appear to be having more success just using BPP. But am I losing out by just doing bpp for everything?

[ballyhoo]

as a follow to my above post, I really might have some significant issue here with PI pre-processing.  Maybe it is in a calibration frame. The reason why I think so, s running BPP,  I am still running into problems.  I did not use masters for BPP.  if anyone can help me out,  I am linking a sample of each calib frame.

https://www.dropbox.com/s/rlpdq2h4wmtdzbm/dark%20120_013.fit?dl=0                     (DARK)

https://www.dropbox.com/s/6p8m6aw38lrieyp/bias%20%20120_002.fit?dl=0                (BIAS)

https://www.dropbox.com/s/zs0mjxeepp5lnbb/FLAT.fit?dl=0                                        (FLAT)

[bulrichl]

Yes, the problem are the settings for the light frame calibration. If you enable the section 'MasterBias' and select the MB, it will be subtracted from the light frames (step 1). Then, if you select the section 'Master Dark' with 'Calibrate' disabled, the MB will not be subtracted from the MD, so the uncalibrated MD will be subtracted from the light frame (step 2). That means both MB and MD are subtracted from the light frames, and this is not correct.

Correct settings:
Section 'Master Bias' disabled.
Section 'Master Dark' enabled, MD selected, both options 'Calibrate' and 'Optimize' disabled.
Section 'Master Flat' enabled, MF selected, option 'Calibrate' disabled

With these settings, only the MasterDark is subtracted from the light frames, then the result is divided by the MasterFlat and multiplied by mean(MF).

Bernd
 

[ballyhoo]

but with the same results now with BPP I am not sure what is going on.

[dave_galera]

Just been looking at the bias, darks and flats....bias and darks look OK but the flat looks like this if you zoom in:

To me this is not what I would expect so I think you flats are up the spout

[dave_galera]

Which is why you image probably looks like this:

To prove it try the calibrating the lights WITHOUT the flat

BTW Your darks are not dark they have light leakage on the right hand side