Problem with DSLR Workflow

[cdavid]

Hi, I have been following the workflow as described.  Last year I had no issues, but it seems that now I am having troubles.  I am creating my MasterDarks, MasterFlats with Bias Calibration as is described.  When I calibrate my light frames, debayer, and then register everything seems ok.  But when I go to integrate I am getting an Error and it fails due to inability to calculate noise....it gives me a warning that it cannot converge the MRS Error.

Not sure what is going on, any ideas?

Thanks
Carlos

[astropixel]

I see this frequently, but I am told that it is not critical, and I believe it may be related to calibration. I find that with sufficient bias darks and flats to make the master frames this error doesn't appear.

I have a set of recent images that virtually all threw this error. The finished image is fine.

[cdavid]

Hi, thanks for the reply.  I have also but now the entire process stops and will not integrate the images due to "error unable to calculate the noise"

not sure what to do.

Carlos

[Juan Conejero]

Hi Carlos,

"error unable to calculate the noise"

This problem has been fixed in the new version of the ImageIntegration module. We'll release it today as an update for all platforms.

but now the entire process stops and will not integrate the images due to "error unable to calculate the noise"

However, this shouldn't happen, and I haven't seen it before. Noise evaluation failure should never stop the integration process. Are you sure there isn't anything else in your data set which is causing this problem?

[cdavid]

I really don't know anymore....I just went through the process again step by step carefully according to vincents outline.  When I finally get to the integration of my calibrated-debayered-registered lights it quits due to error.  It tells me there is no signal in the image that it is empty??

Also when calibrating the lights, I am getting a warning message that there is no correlation between my darks and lights?  I don't understand this, they were both taken at the same ISO and exposure time.

I have the raw preferences set to no white balances, no black point correction, and set to create raw bayer image (not the cfa option).

Any ideas
Thanks Juan

Carlos

[zvrastil]

Hi Carlos,

verify if your calibrated images contain some information at all (for example with STF or ImageStatistics). I recently reported problem with ImageCalibration which produces completely black calibrated images as long as I provide RAW files to ImageCalibration. If I convert them to FIT first, the image is calibrated correctly. If you calibrated images are totally black (all pixels = 0), it could explain why processing fails.

regards, Zbynek

[cdavid]

Thanks, I did that....they are not completely without signal.....but mainly a blur of noise in red and green colors.  I will try converting the RAWS to Fits first and see if that helps
Thanks again
CArlos

[cdavid]

Well I have tried various methods.  Using my light subs as Fits instead of DSLR RAW.  I tried just making a dark and flat without calibration and later calibrating when I integrated the lights.  Every method fails.  During the calibration I get a warning that my dark and target image have no correlation.

Then after debayer and registration I try to integrate and it quits immediately with the message of "little or insignificant image, empty image?"

Not really sure what I am doing wrong here.  I am going to try preprocessing in another program to make sure my darks are not corrupted somehow...even though the master dark in Pix in sight does look right with the amp glow in the corner, etc.

Thanks again for all your suggestions.
Carlos

[Enzo De Bernardini]

Carlos, si hablás español (supongo que si) te recomiendo chequear los seteos de DSLR_RAW, estar seguro que tanto los lights como las tomas de calibración se abren con los mismos parámetros, que la integración de las tomas de calibración es la adecuada y los seteos en ImageCalibration también son adecuados. Para ello, quizá te sirva de guía el siguiente tutorial que hice algún tiempo:

http://astrosurf.com/astronosur/pixinsight/preprocessing-1.htm

Espero te sea de utilidad y resuelvas estos problemas.

Saludos,

Enzo.

[astropixel]

I have seen the 'empty' error in the past and found it to be related to improper application of the calibration process. Exactly, I can't remember, but following Vicent's instructions in the expanded version linked to the dslr_workflow - no further problems.

[cdavid]

Gracias Enzo, yes I speak and read Spanish but my writing is poor so I will respond in English. Thank you for your tutorial it is very good. This is however what I was doing with the exception of calibrating my masterdark during it's creation. I am still getting this error. What is funny is I had no problem calibrating and stacking with Nebulosity. I just downloaded the update to image integration and will try again. Perhaps the update will fix this.

Astropixel....I agree but I have been following vin cents instructions to the letter and still having problems.

Thanks again

[cdavid]

Utter frustration.  I updated the Integration tool earlier tonight.  I went through and attempted to pre-process my images again.  I followed Enzo's tutorial to the letter.

I get these messages during the light calibration:

Calibrating target frame 29 of 41
Loading target frame:
/Volumes/FireLite 100 GB/Wizard Nebula NGC 7380/_MG_2306.CR2
Timestamp: Wed Aug 31 03:04:34 2011
Camera: Canon EOS DIGITAL REBEL XS
Owner: Carlos David
ISO speed: 1600
Shutter: 361.0 sec
Aperture: f/inf
Focal length: 0.0 mm
Embedded ICC profile: no
Number of raw images: 1
Thumb size:  1936 x 1288
Full size:   3948 x 2622
Image size:  3906 x 2602
Output size: 2602 x 3906
Raw colors: 3
Filter pattern: RGGBRGGBRGGBRGGB
Daylight multipliers: 2.362489 0.919100 1.182562
Camera multipliers: 1827.000000 1024.000000 1211.000000 1024.000000

Invoking: dcraw -D -k 0 -t 0 -o 0 -4
Decoding Canon EOS DIGITAL REBEL XS file (3906x2602 pixels, ISO=1600, Exposure=361.00s): 100%
Loading raw image: 100%

Writing output file: /Users/Carlos/Desktop/Wizard Enzo Recs/Calibrated lights/_MG_2306_c.fit
Dark scaling factors:
k0 = 0.000
** Warning: No correlation between the master dark and target frames.
k1 = 1.312
k2 = 0.636
Writing FITS image: 32-bit floating point, 3 channel(s), 3906x2602 pixels: 100%


and earlier during the Flats calibration I received these errors:
Opening files:
/Users/Carlos/Desktop/Wizard Enzo Recs/Calibrated Flats/_MG_2339_c.fit
Reading FITS: 32-bit floating point, 3 channel(s), 3906x2602 pixels: 100%
MRS noise evaluation: |
** Warning: No convergence in MRS noise evaluation routine - using k-sigma noise estimate.|
** Warning: No convergence in MRS noise evaluation routine - using k-sigma noise estimate./
** Warning: No convergence in MRS noise evaluation routine - using k-sigma noise estimate.done.
Weight          :     1.00000    1.00000    1.00000


Finally, when I went to star align after debayering it fails.  It can't find star matches.

I just don't know what is going on....I have been using pixinsight for about a year and have never had these issues.

Carlos

[zvrastil]

Hi Carlos,

try to switch off dark optimization, if you did not yet. I understand you have amplifier glow in your dark frames (you mention it in one post). As far as I know, dark optimization does not work when amp glow is there. Without optimization, your darks must match lights in exposure time, iso and temperature.

regards, Zbynek

[astropixel]

Hi Carlos. I hope that works for you. Out of curiosity have these calibration files been previously calibrated by another program?

[cdavid]

Thanks for the replies...I will try that next.  Astropixel- nope, in fact I have freshly converted the RAW files to FITS each time to make sure they weren't somehow corrupted.  I will keep you posted.

Thanks again
Carlos