taskGNU Astronomy Utilities - Tasks: task #13557, WCS meta-data of many images in a...

 
 

You are not allowed to post comments on this tracker with your current authentication level.

task #13557: WCS meta-data of many images in a table

Submitter:  Mohammad Akhlaghi <makhlaghi>
Submitted:  Mon 20 Apr 2015 06:14:33 AM UTC
 
Should Start On:  Sun 19 Apr 2015 03:00:00 PM UTC Should be Finished on:  Sun 19 Apr 2015 03:00:00 PM UTC
Category:  All Gnuastro Priority:  7 - High
Item Group:  New feature Status:  Postponed
Privacy:  Public Percent Complete:  0%
Assigned to:  None Open/Closed:  Open
Effort:  0.00

Wed 19 Sep 2018 01:03:44 PM UTC, comment #2: 

The IVOA MOC (Multi-Order Coverage map) standard is also apparently a good way to efficiently find overlaps between images on the Sky for this task. We should look into it.

Maybe the table doesn't have to be raw WCS information, but instead use the HEALPix sky tessellation algorithm for making the coverage of each image on the celestial sphere.

Mohammad Akhlaghi <makhlaghi>
Project Administrator
Tue 18 Sep 2018 05:16:35 PM UTC, comment #1: 

Updated statement: In some contexts it is necessary for Gnuastro's programs to load the WCS meta-data of many files. One example can be using Crop's WCS-mode cropping which can stitch many images if the object lies on the border of multiple images. Another can be MakeCatalog when it wants to map the segmentation map on a region in the border of another image (not yet fully implemented).

In surveys (where the reduced image is broken into tiles), the WCS structure is already matched between all the images. With only certain elements differing. Having to parse/convert the FITS keywords of all the files in every run of every program can be very inefficient or even impractical (especially on large surveys like LSST). To solve this problem, we can parse all the images once and make a FITS table. In its header, we can write a reference WCS structure and in each row, we can have the values that need to be changed for each input along with the file/extension of each. Afterwards, any program can only to translate one set of FITS keywords into an internal WCS structure, and parse the table to find which image it needs (has an overlap with the region it wants) for its higher-level processing.

Before writing, the table can also be sorted in a specific way to help user programs find the images they want more efficiently (without having to check all the table).

This can be implemented as a library feature to be used in all programs and also by users in their own programs.

Mohammad Akhlaghi <makhlaghi>
Project Administrator
Mon 20 Apr 2015 06:14:33 AM UTC, original submission:  

Add an option to save/read the required meta-data of each input image in an ASCII file. Such that all the information in the inputimgs structure (in main.h) is included in it. Such a file can become important when the images are always the same and there are lots of requests. The reason a binary might be necessary is that it includes the wcsprm structure of WCSLIB. We certainly don’t want to run the WCSLIB settings each time (they are not thread safe and can be slow). The good thing with an ASCII file is that it is transparent.

Mohammad Akhlaghi <makhlaghi>
Project Administrator

 

(Note: upload size limit is set to 16384 kB, after insertion of the required escape characters.)

No files currently attached

 

Depends on the following items: None found

Items that depend on this one: None found

 

Carbon-Copy List
  • -email is unavailable- added by makhlaghi (Submitted the item)
  •  

    There are 0 votes so far. Votes easily highlight which items people would like to see resolved in priority, independently of the priority of the item set by tracker managers.

     

     

     

    Follow 3 latest changes.

    Date Changed by Updated Field Previous Value => Replaced by
    2018-09-18 makhlaghi CategoryCrop All Gnuastro
        Priority1 - Later 7 - High
        SummaryRead image meta-data from file WCS meta-data of many images in a table

    Back to the top


    Powered by Savane 3.9