Welcome to iraf.net Tuesday, May 14 2024 @ 07:46 PM GMT


 Forum Index > Help Desk > Applications New Topic Post Reply
 yet another wregister question
   
aoxsic
 03/13/2007 06:36PM (Read 3628 times)  
++---
Junior

Status: offline


Registered: 03/13/2007
Posts: 16
Hello!First, some background:
I'm writing a reduction script(optical and near IR dithered data, non-interactive) that should be pretty automatic and robust. I've come across images who have no WCS in their headers. I could use xregister to align those, but I don't want to sit and click on a star in every image. I want the alignment to happen automatically. I tried to avoid WCS altogether, so I decided to use xregister but create all the input it needs automatically. I've written a procedure that runs sextractor on one frame, and gets 3 reference stars that are relatively close to the center(to make sure they'll be in every frame). Then I do a very preliminary guess at the initial shifts by simply looking at the RA and DEC of the reference file and comparing it to the rest of the frames(I know the pixel per arcsecs for that instrument). But it just occured to me that those shifts would surely be wrong if the frames were taken near the poles since I won't know how many arcsecs per pixel that would make. I ran some test, and it almost works this way but the shifted images are not quite aligned, they jump around a bit. However, I'm sure it "almost works" for this set of data while it will completely fail for another. I can't have that, it should work for everything.End of background.So, then I'm back to square one - I need shifts. I guess that's exactly what wregister was written for, but the IRAF help for that task is very cryptic to me, I don't understand what it wants from me! Surely I can't simply add all the missing WCS keywords, can I? And how would I know what they are? What values go with some particular image? Where is that described?So, please, I need some ideas on how to create a WCS in a fits header. Point me to some manual, maybe? Any help, manual links, etc. much appreciated.

 
Profile Email
 Quote
fitz
 03/13/2007 06:36PM  
AAAAA
Admin

Status: offline


Registered: 09/30/2005
Posts: 4040
If you want to avoid the WCS altogether then have a look at the XYXYMATCH task as a way to match coordinate lists (e.g. the SExtractor catalogs for each image, see also STARFIND in the system). What WREGISTER does is similar except that it uses the WCSXYMATCH task to do the registration using the WCS, and then calls GEOMAP/GEOTRAN to transform the images. If all you want to do is apply a shift then XYXYMATCH will give you that, IMLINTRAN can do the transformation.Computing a real WCS solution from this match requires you do a full plate solution and match the xy coords to a reference catalog of positions. There are ways to do that as well but it sounds like it's beyond what your script needs to do.-Mike

 
Profile Email
 Quote
aoxsic
 03/13/2007 06:36PM  
++---
Junior

Status: offline


Registered: 03/13/2007
Posts: 16
Thank you, I will try xyxymatch. I hope it doesn't just do shifts, but handles rotation and scaling as well. If it does, I'm all set! Thanks a bunch!!
[color=red:1fa398664e][Moderator's Note: scale and rotation are indeed handled] [/color:1fa398664e]

 
Profile Email
 Quote
aoxsic
 03/13/2007 06:36PM  
++---
Junior

Status: offline


Registered: 03/13/2007
Posts: 16
Just wanted to say: Yay, it works! Thank you, Mike!

 
Profile Email
 Quote
   
Content generated in: 0.14 seconds
New Topic Post Reply

Normal Topic Normal Topic
Sticky Topic Sticky Topic
Locked Topic Locked Topic
New Post New Post
Sticky Topic W/ New Post Sticky Topic W/ New Post
Locked Topic W/ New Post Locked Topic W/ New Post
View Anonymous Posts 
Anonymous users can post 
Filtered HTML Allowed 
Censored Content 
dog allergies remedies cialis 20 mg chilblain remedies


Privacy Policy
Terms of Use

User Functions

Login