Imaging

rascil.processing_components.imaging.base Module

Functions that aid fourier transform processing. These are built on top of the core functions in processing_components.fourier_transforms.

The measurement equation for a sufficently narrow field of view interferometer is:

\[V(u,v,w) =\int I(l,m) e^{-2 \pi j (ul+vm)} dl dm\]

The measurement equation for a wide field of view interferometer is:

\[V(u,v,w) =\int \frac{I(l,m)}{\sqrt{1-l^2-m^2}} e^{-2 \pi j (ul+vm + w(\sqrt{1-l^2-m^2}-1))} dl dm\]

This and related modules contain various approachs for dealing with the wide-field problem where the extra phase term in the Fourier transform cannot be ignored.

Functions

`shift_vis_to_image`(vis, im[, tangent, inverse])	Shift visibility in place to the phase centre of the Image
`normalise_sumwt`(im, sumwt[, min_weight, ...])	normalise out the sum of weights
`predict_awprojection`(vis, model[, gcfcf])	Predict using convolutional degridding and an AW kernel
`invert_awprojection`(vis, im[, dopsf, ...])	Invert using convolutional degridding and an AW kernel
`create_image_from_visibility`(vis, **kwargs)	Make an empty image from params and Visibility
`advise_wide_field`(vis[, delA, ...])	Advise on parameters for wide field imaging.
`visibility_recentre`(uvw, dl, dm)	Compensate for kernel re-centering - see w_kernel_function.
`fill_vis_for_psf`(svis)	Fill the visibility for calculation of PSF

rascil.processing_components.imaging.dft Module

Functions that aid fourier transform processing. These are built on top of the core functions in processing_components.fourier_transforms.

The measurement equation for a sufficently narrow field of view interferometer is:

\[V(u,v,w) =\int I(l,m) e^{-2 \pi j (ul+vm)} dl dm\]

The measurement equation for a wide field of view interferometer is:

\[V(u,v,w) =\int \frac{I(l,m)}{\sqrt{1-l^2-m^2}} e^{-2 \pi j (ul+vm + w(\sqrt{1-l^2-m^2}-1))} dl dm\]

This and related modules contain various approachs for dealing with the wide-field problem where the extra phase term in the Fourier transform cannot be ignored.

Functions

`dft_skycomponent_visibility`(vis, sc, **kwargs)	DFT to get the visibility from a SkyComponent, for Visibility
`idft_visibility_skycomponent`(vis, sc)	Inverse DFT a SkyComponent from Visibility

rascil.processing_components.imaging.imaging_params Module

Functions

`get_rowmap`(col[, ucol])	Map to unique cols
`get_polarisation_map`(vis[, im])	Get the mapping of visibility polarisations to image polarisations
`get_frequency_map`(vis[, im])	Map channels from visibilities to image

rascil.processing_components.imaging.ng Module

Functions that implement prediction of and imaging from visibilities using the nifty gridder (DUCC version).

https://gitlab.mpcdf.mpg.de/mtr/ducc.git

This performs all necessary w term corrections, to high precision.

Note that nifty gridder doesn’t like some null data such as all w = 0 and do_wstacking=True. Also true of the visibilities.

Functions

`predict_ng`(bvis, model, **kwargs)	Predict using convolutional degridding.
`invert_ng`(bvis, model[, dopsf, normalise])	Invert using nifty-gridder module

rascil.processing_components.imaging.primary_beams Module

Functions to create primary beam and voltage pattern models

Functions

`set_pb_header`(pb[, use_local])	Fill in PB header correctly for local coordinates.
`create_pb`(model[, telescope, ...])	Create an image containing the primary beam for a number of cases
`create_pb_generic`(model[, pointingcentre, ...])	Create a generic analytical model of the primary beam
`create_vp`([model, telescope, ...])	Create an image containing the dish voltage pattern for a number of cases
`create_vp_generic`(model[, pointingcentre, ...])	Create a generic analytical model of the voltage pattern
`create_vp_generic_numeric`(model[, ...])	Make an image like model and fill it with an analytical model of the primary beam
`create_low_test_beam`(model[, use_local, azel])	Create a test power beam for LOW
`create_low_test_vp`(model[, use_local, azel])	Create a test voltage beam for LOW
`create_mid_allsky`([frequency, npixel, cellsize])	Approximate all sky MID beam
`convert_azelvp_to_radec`(vp, im, pa)	Convert AZELGEO image to image coords at specific parallactic angle
`normalise_vp`(vp)	Normalise the vp in place so that the peak gain on axis for parallel pols is equal

rascil.processing_components.imaging.weighting Module

Functions that aid weighting the visibility data prior to imaging.

There are two classes of functions:

Changing the weight dependent on noise level or sample density or a combination
Tapering the weihght spatially to avoid effects of sharp edges or to emphasize a given scale size in the image

Functions

`weight_visibility`(vis, model[, weighting, ...])	Weight the visibility data
`taper_visibility_gaussian`(vis[, beam])	Taper the visibility weights
`taper_visibility_tukey`(vis[, tukey])	Taper the visibility weights

rascil.processing_components.imaging.wg Module

Functions that implement prediction of and imaging from visibilities using the GPU-based gridder (WAGG version), https://gitlab.com/ska-telescope/sdp/ska-gridder-nifty-cuda

Currently the python wrapper of the GPU gridder is available in a branch, https://gitlab.com/ska-telescope/sdp/ska-gridder-nifty-cuda/-/tree/sim-874-python-wrapper

This performs all necessary w term corrections, to high precision.

Functions

`predict_wg`(bvis, model, **kwargs)	Predict using convolutional degridding.
`invert_wg`(bvis, model[, dopsf, normalise])	Invert using GPU-based WAGG nifty-gridder module