tmolteno / disko Goto Github PK

View Code? Open in Web Editor NEW

2.0 2.0 2.0 1.44 MB

Discrete Sky Operator (DiSkO) Aperture Synthesis Radio Imaging

License: GNU General Public License v3.0

Makefile 2.68% Python 96.83% Dockerfile 0.18% Shell 0.31%

disko's People

Contributors

Stargazers

Watchers

Forkers

mfkiwl bennahugo

disko's Issues

Add a feature to disko_draw to allow manual specification of the pixel range

Add a feature to disko_draw to allow manual specification of the pixel range:

This would allow fixed min and max scales for images to create movies with (for example).

disko_draw --min 0.0 --max 16 ...

Dumping FITS files from json writes fits files without a valid WCS in the header

This is only an issue when going from json, not from ms.

The images are the same, but the WCS is invalid for the former

SIMPLE  =                    T / conforms to FITS standard                      
BITPIX  =                  -32 / array data type                                
NAXIS   =                    2 / number of array dimensions                     
NAXIS1  =                 2000                                                  
NAXIS2  =                 2000                                                  
ORIGIN  = 'DiSkO   '           / L-2 Regularizing imager written by Tim Molteno 
CRPIX1  =               1001.0                                                  
CDELT1  = -0.06484999999999999                                                  
CRPIX2  =               1001.0                                                  
CDELT2  =  0.06484999999999999                                                  
COMMENT DiSkO: None                                                             
END

vs. going from MS

SIMPLE  =                    T / conforms to FITS standard                      
BITPIX  =                  -32 / array data type                                
NAXIS   =                    2 / number of array dimensions                     
NAXIS1  =                 2000                                                  
NAXIS2  =                 2000                                                  
ORIGIN  = 'DiSkO   '           / L-2 Regularizing imager written by Tim Molteno 
CRPIX1  =               1001.0                                                  
CDELT1  = -0.06484999999999999                                                  
CRPIX2  =               1001.0                                                  
CDELT2  =  0.06484999999999999                                                  
CTYPE1  = 'RA---SIN'           / Right ascension angle cosine                   
CRVAL1  =    355.1778032271041                                                  
CUNIT1  = 'deg     '                                                            
CTYPE2  = 'DEC--SIN'           / Declination angle cosine                       
CRVAL2  =   -45.97593928027329                                                  
CUNIT2  = 'deg     '                                                            
CTYPE3  = 'FREQ    '                                                            
CRPIX3  =                  1.0                                                  
CRVAL3  = '1575420000.0'                                                        
CDELT3  =      10026896.158854                                                  
CUNIT3  = 'Hz      '                                                            
EQUINOX = '2000.   '                                                            
DATE-OBS= '5173064422.535001'                                                   
BTYPE   = 'Intensity'                                                           
COMMENT DiSkO: None                                                             
END

Convert processing to U,V,W stored in meters

Currently in disko, it is assumed that U,V,W are stored in wavelengths. This is a problem for managing multiple frequency bands.

Change the disko object to use meters
Modify the code for handling tart to behave.

Disko overplotting coordinate system does not work for anything other than Zenith projection

It looks like the coordinate system is not using WCS to work out where on the sphere to place things. The GNSS plots has no respect for PHASE_DIR
Not phase steered, single field ms:

Phase steered, single field ms:

Move to tox for testing

Everything else seems broken. Particularly with tests_require which isn't working any more.

Get sources from json and interpolate them over the MSv2 timestamps

Currently the sources overplotting only works when you go from JSON

I need to somehow get an NN temporal interpolation scheme going to read data from MS and sources from JSON an then overplot sources based on the mean timestamp of the selected field from a multifield MS

Co-ordinate aware sphere object with Persistence

This issue is to track the development of a new co-ordinate aware sphere. The sphere should be able to manage it's own transformations between elevation-azimuth as well as celestial co-ordinates.

Every pixel has co-ordinates
Introduce the concept of the phase centre of the sphere. This is the pixel that represents the phase centre, and it will have celestial and el_az coordinates.
Add methods to transform between co-ordinates celestial_to_elaz, elaz_to_celestial
Create a storage format for spheres that can be reused. This is likely to be vtk.
Shift drawing to a new utility (disko_plot) that draws over sphere's. Also exports as FITS e.t.c.

Issue #7 is relevant here. This will allow plotting of sources in either coordinate system (satellites, or celestial objects)

Field and DDID selector switches

Currently field ID reads phase center for field id 0. DDID is also not flexible enough

Fix incoming

Default behaviour should make nice images

Right now, disko should have some sensible defaults for creating moderately good TART images.

The following might make sense...

disko --tikhonov --alplha 0.02

Convention of MS UVW are displaying upside down in DISKO

This is because DISKO I think is using the wrong handedness convention (or at least it is only self-consistent with telescope mode UVW).
We need to be consistent with accepted NRAO convention to read correctly from MS.

Will fix

Test: test_imaging_vs_natural failing

This test is failing since DiSkO moved to an augmented (re, im) vector rather than complex vectors for the telescope operator. Two approaches are possible:

Revert to complex operators. This is preferred, however quite a few packages don't support complex operators.
Fix the projection into the natural basis...

tmolteno / disko Goto Github PK

disko's People

Contributors

Stargazers

Watchers

Forkers

disko's Issues

Recommend Projects

Recommend Topics

Recommend Org

Jobs