Post Reply 
 
Thread Rating:
  • 0 Votes - 0 Average
  • 1
  • 2
  • 3
  • 4
  • 5
All zeros in SWarped image
10-16-2006, 22:54
Post: #1
All zeros in SWarped image
When I am stacking images using SWarp with either 2.15.7 or 2.16, I get all zeros in the stacked image. I thought it had something to do with an error I was getting (see my reply in the thread "weighting factor set to 1"), although I am not so sure. I tried running a similar script with SWarp 2.15, and it worked! The parameters I used are below.

-IMAGEOUT_NAME getcenterstack1.fits -WEIGHTOUT_NAME getcenterstack1.weight.fits -COMBINE_TYPE AVERAGE -MEM_MAX 1024 -WEIGHT_TYPE MAP_WEIGHT -SUBTRACT_BACK N -WEIGHT_IMAGE swircbpm.fits

I still get the "WARNING: Null or negative global weighting factor:defaulted to 1", but the resulting image is what I would expect. Any reason why 2.15 would work better? Let me know if you want my full script or any images. I would love to get it to work with 2.16, because I have ~150 input images, and 2.15 can't handle that many in one command. Thanks!!!
Find all posts by this user
Quote this message in a reply
10-16-2006, 23:25
Post: #2
RE: All zeros in SWarped image
Check that, I no longer get the warning, so perhaps they are related...
Find all posts by this user
Quote this message in a reply
10-17-2006, 10:51
Post: #3
RE: All zeros in SWarped image
Well, that warning is coming the fact that your weight-map is full of zeroes.
When you swarp something with a weightmap consisting of zeros, well, the output
is also zero....

cheers
henry
Visit this user's website Find all posts by this user
Quote this message in a reply
10-17-2006, 10:52
Post: #4
RE: All zeros in SWarped image
-- you can obviously test this by swarping without the weight map.
Visit this user's website Find all posts by this user
Quote this message in a reply
10-17-2006, 18:09
Post: #5
RE: All zeros in SWarped image
The weight map is not all zeros, it is mostly ones.  I did try SWarping without the weightmap, and while the warning did not occur, the end result was still the same (zero image).

To try to diagnose the problem, I ran 4 different scenarios, trying to SWarp one image with and without the weightmap, using v2.15.7 and v2.15 of SWarp.  With or without the weightmap, v2.15.7 produced an output of all zeros, while with or without the weightmap, v2.15 produced an acceptable result.  This would lead me to believe that the actual weightmap is not the problem.  Is there any reason why v2.15 would do this right and 2.15.7 would not?  Here is the command line input:

swarp bootes1.6281s.fits -IMAGEOUT_NAME swarp.6281.fits -WEIGHTOUT_NAME swarp.6281.weight.fits -CENTER_TYPE MANUAL -CENTER 14:25:04.65,35:37:29.11 -IMAGE_SIZE 3067,3001 -PIXEL_SCALE 0.15 -PIXELSCALE_TYPE MANUAL -COMBINE_TYPE AVERAGE -MEM_MAX 1024 -WEIGHT_TYPE MAP_WEIGHT -WEIGHT_IMAGE swircbpm.fits -SUBTRACT_BACK N


I had trouble uploading attachments, so I put the relavent files on my website at:

http://stevenf.asu.edu/swarp

if anyone wants to try to repeat me error.

One other quick question:  the last time I used SWarp was in May, and it had a keyword ROTATION, which I had used to make sure it didn't rotate the images.  That keyword doesn't exist in eith 2.15, 2.15.7 or 2.16, and thus (when I get SWarp to work), some of my images end up rotated.  Is there some other way to tell SWarp not to rotate the output?  Thanks so much for any help!

- Steven Finkelstein
Find all posts by this user
Quote this message in a reply
10-24-2006, 20:01 (This post was last modified: 10-24-2006 20:01 by Emmanuel Bertin.)
Post: #6
RE: All zeros in SWarped image
I will check this tomorrow when I am back.
Uh I do not remember this ROTATION keyword!!! Do you mean the CROTA2 in the FITS header?
- Emmanuel.
Visit this user's website Find all posts by this user
Quote this message in a reply
10-26-2006, 20:22
Post: #7
RE: All zeros in SWarped image
The keyword was rotation - we were using a version of SWarp onsite at a telescope, but we haven't touched the reduction since then, about 5 months ago, so the details are foggy in my head. But I think we can get around that, my main concern was why an identical process worked with version 2.15 and not 2.15.7 or 2.16, thanks!
Find all posts by this user
Quote this message in a reply
11-10-2006, 16:53
Post: #8
RE: All zeros in SWarped image
Strange. I just ran SWarp 2.16.0 directly from your script, both with the 32 and the 64 bit versions, multithreaded and single-threaded, and I get a correct image without the zeros you mention. I also ran it through Valgrind, and got absolutely no warning. Sorry, what else can I do? I am puzzled.
- Emmanuel.
Visit this user's website Find all posts by this user
Quote this message in a reply
09-21-2008, 14:25
Post: #9
RE: All zeros in SWarped image
i have the same problem, when running swarp 2.17 with SDSS images, i get only all-zero output.
Find all posts by this user
Quote this message in a reply
Post Reply 


Forum Jump:


User(s) browsing this thread: 1 Guest(s)