`virgil.scenes`

Synthetic truth images for testing image reconstructions: simulate data from a known scene with your own coverage and noise before trusting an image of real data.

Synthetic truth images for testing image reconstructions.

Each function returns a unit-sum (npix, npix) array in the virgil image orientation (East left, North up, centre at the middle of the pixel grid, as in render), ready for Image.from_brightness. Position angles run North to East. The ring and spiral shapes are inspired by the training scenes of Jonah Goldfine's frito (https://github.com/JonahDG/frito), re-implemented here in virgil conventions.

`ring(npix, pixel_scale_mas, radius_mas, width_mas, inc_deg=0.0, pa_deg=0.0, asymmetry=0.0, asymmetry_pa_deg=0.0)`

Gaussian-profile ring, optionally inclined and azimuthally modulated.

Parameters:

Name	Type	Description	Default
`npix`	`int`	Number of pixels on a side.	required
`pixel_scale_mas`	`float`	Pixel size in milliarcseconds.	required
`radius_mas`	`float`	Radius of the ring's peak in milliarcseconds (the semi-major axis when inclined).	required
`width_mas`	`float`	Standard deviation of the radial profile in milliarcseconds.	required
`inc_deg`	`float`	Inclination in degrees; 0 is face-on, and the minor axis is `cos(inc)` times the major axis.	`0.0`
`pa_deg`	`float`	Position angle of the major axis, North to East, in degrees.	`0.0`
`asymmetry`	`float`	Amplitude (0 to 1) of the modulation `1 + asymmetry * cos(PA - asymmetry_pa_deg)` of the brightness around the ring, where PA is the on-sky position angle of a pixel.	`0.0`
`asymmetry_pa_deg`	`float`	Position angle, North to East, of the brightest part of the ring, in degrees.	`0.0`

Returns:

Type	Description
`(Array, shape(npix, npix))`	The ring, summing to one.

`spiral(npix, pixel_scale_mas, step_mas, width_mas, turns=2.0, pa_deg=0.0, fade_mas=None)`

Archimedean spiral, like the dust pinwheels of WR 104 and WR 137.

The arm is r = step_mas * theta / (2 pi), starting at the centre and winding towards increasing position angle as it moves outwards.

Parameters:

Name	Type	Description	Default
`npix`	`int`	Number of pixels on a side.	required
`pixel_scale_mas`	`float`	Pixel size in milliarcseconds.	required
`step_mas`	`float`	Radial distance between successive turns, in milliarcseconds.	required
`width_mas`	`float`	Standard deviation of the Gaussian cross-section, measured along the radial direction, in milliarcseconds.	required
`turns`	`float`	Number of turns of the arm, which ends in a rounded cap.	`2.0`
`pa_deg`	`float`	Position angle, North to East, of the start of the arm, in degrees.	`0.0`
`fade_mas`	`float`	If given, the brightness falls off as `exp(-r / fade_mas)`.	`None`

Returns:

Type	Description
`(Array, shape(npix, npix))`	The spiral, summing to one.

`gaussian_blob(npix, pixel_scale_mas, sigma_mas, dra=0.0, ddec=0.0)`

Circular Gaussian, e.g. a clump or a resolved companion.

Parameters:

Name	Type	Description	Default
`npix`	`int`	Number of pixels on a side.	required
`pixel_scale_mas`	`float`	Pixel size in milliarcseconds.	required
`sigma_mas`	`float`	Standard deviation in milliarcseconds.	required
`dra`	`float`	Offset of the centre from the image centre in milliarcseconds, positive to the East and North.	`0.0`
`ddec`	`float`	Offset of the centre from the image centre in milliarcseconds, positive to the East and North.	`0.0`

Returns:

Type	Description
`(Array, shape(npix, npix))`	The blob, summing to one.