Bug 414155

Summary:	EKOS produces FITS files with inconsistent values (CDELT1, CDELT2)
Product:	[Applications] kstars	Reporter:	han.k <han.k>
Component:	general	Assignee:	Jasem Mutlaq <mutlaqja>
Status:	RESOLVED DOWNSTREAM
Severity:	normal
Priority:	NOR
Version First Reported In:	unspecified
Target Milestone:	---
Platform:	Other
OS:	Linux
Latest Commit:		Version Fixed/Implemented In:
Sentry Crash Report:

Description han.k 2019-11-14 21:25:26 UTC

SUMMARY

For FITS files written by EKOS, the header values CDELT1,CDELT2 values don't match up with the focal length and pixel size. See link to image. Same for an other image see forum:

https://indilib.org/forum/ekos/5901-astap-astrometry-integration.html?start=48#45813

From the header:

Focal length 560 mm
Pixel size 3.8 micrometer. Pixel size 1.4 arcsec/pixel equals 0.000391 degrees/pixel
Dimensions 4656 x 3520 => Image size is 1.82 x 1.37 degrees
CDELT1,2 in header 0.000777 degrees/pixel but should be 0.000391 degrees/pixel

What's going wrong ?


image link:
dl.free.fr/getfile.pl?file=/gAfdJ7sV


STEPS TO REPRODUCE
1. 
2. 
3. 

OBSERVED RESULT


EXPECTED RESULT


SOFTWARE/OS VERSIONS
Windows: 
macOS: 
Linux/KDE Plasma: 
(available in About System)
KDE Plasma Version: 
KDE Frameworks Version: 
Qt Version: 

ADDITIONAL INFORMATION

Comment 1 Jasem Mutlaq 2019-11-14 22:22:06 UTC

was this image captured on the guide scope?

Comment 2 Jasem Mutlaq 2019-11-15 08:39:31 UTC

So this is an issue with INDI, not Ekos/KStars. INDI generates FITS, KStars is just a client.

I *believe* it can happen when users capture in guide scope, but the INDI::CCD still utilizes primary focal length for CDELT calculations.

Comment 3 Jasem Mutlaq 2019-11-25 10:18:21 UTC

This issue was fixed in INDI. So closing this unless it is KStars related.