psfsim.opticspsf

Optics objects.

Classes

GeometricOptics

Geometric optics object.

Functions

`altgriddata`(points, values, xi)	Quadratic fitting function.
`fit_linfunc`(f)	Helper function to do a linear fit: f = a + bx + cy.
`compute_jacobian`(u[, dx, dy])	Computes the Jacobian for entrance --> exit pupil.

Module Contents

altgriddata(points, values, xi)[source]

Quadratic fitting function.

This is a substitute for scipy.interpolate.griddata when we have 5 points and want to fit a quadratic, minus the x^2-y^2 term.

Paramters

pointsnp.ndarray of float: The coordinates of the 5 points where we have values; shape (5, 2).
valuesnp.ndarray of flat: The values to interpolate; shape (5,).
xi(float, float): The location to interpolate to.

returns:: The interpolated value.
rtype:: float

fit_linfunc(f)[source]

Helper function to do a linear fit: f = a + b*x + c*y.

Excludes nans. Here x and y start at 0.

Parameters:: f (np.ndarray of float) – The function to fit. Should be a 2D array, f[y, x].
Returns:: coefs – The coefficients of the fit. Length 4: a, b, c, rmserr.
Return type:: np.ndarray of float

compute_jacobian(u, dx=1.0, dy=1.0)[source]

Computes the Jacobian for entrance –> exit pupil.

Parameters:

u (np.ndarray of float) – 3D array of exit pupil positions; u[iy, ix, ic] is the orthographic direction of the outgoing ray in entrance pixel (ix, iy). The components are ic == 0 for the x-component of u and ic == 1 for the y-component of u.
dx (float, optional) – The entrance pupil grid spacings.
dy (float, optional) – The entrance pupil grid spacings.

Returns:

The Jacobian, d(u_x,u_y)_out / d(x,y)_in. Shape is (N, N, 2, 2), where the first 2 axes refer to the shape of u, and the second 2 axes are matrix axes.

Return type:

np.ndarray of float

class GeometricOptics(scanum, scax, scay, wavelength=0.48, use_filter='H', ulen=2048, ray_trace=True, pixelsampling=1.0, a_lanczos=3, cycle=9, mjd=None)[source]

Geometric optics object.

Parameters:

scanum (int) – The SCA number.
scax (float) – The pixel positions on the SCA (in mm, FPA coordinates relative to the SCA center).
scay (float) – The pixel positions on the SCA (in mm, FPA coordinates relative to the SCA center).
wavelength (float, optional) – The vacuum wavelength in microns.
use_filter (str) – The filter as a 1-character string.
ulen (int, optional) – The size of array for pupil sampling.
ray_trace (bool, optional) – Whether to use ray tracing.
pixelsampling (float, optional) – Desired FFT-based output pixel sampling in microns.
cycle (int, optional) – Which cycle to use for the Zernike modes.
mjd (float, optional) – The MJD to use for the optical model.

wavelength[source]

The vacuum wavelength in microns.

Type:: float

scanum[source]

The SCA number.

Type:: int

scax, scay

The pixel positions on the SCA (in mm, FPA coordinates relative to the SCA center).

Type:: float

xan, yan

The field angles in degrees.

Type:: float

use_filter[source]

The filter as a 1-character string.

Type:: str

ulen[source]

The FFT length (and pupil grid size).

Type:: int

ucen, vcen

The orthographic coordinates at the exit pupil of the center of the pupil image.

Type:: float

du[source]

The grid spacing in orthographic coordinates at the exit pupil.

Type:: float

path_difference[source]

The wavefront map in microns. Shape (ulen, ulen).

Type:: np.ndarray of float

rb

The ray trace object.

Type:: psfsim.romantrace.RayBundle

a_lanczos[source]

The order of Lanczos kernel apodization to use for the high-resolution pupil.

Type:: int, optional

__init__()[source]: Constructor.

u_array()[source]: Gets the 2D array of u.

v_array()[source]: Gets the 2D array of v.

compute_distortion_matrix()[source]: Computes the distortion matrix.

compute_determinant()[source]: Determinant of distortion matrix.

load_pupil_mask()[source]: Loads the pupil mask.

path_diff()[source]: Path difference map.

wavelength = 0.48[source]

dsX = 1.0[source]

pupilLength = 19200[source]

focalLength = 8[source]

samplingwidth = 9216.0[source]

scanum[source]

scax[source]

scay[source]

posOut[source]

ulen = 2048[source]

a_lanczos = 3[source]

pupilSampling = 2048[source]

use_filter = 'H'[source]

cycle = 9[source]

mjd = None[source]

perturbations = None[source]

distortionMatrix[source]

determinant[source]

pupil_mask[source]

uvcoefs[source]

ucen[source]

vcen[source]

du[source]

path_difference[source]

u_array()[source]: Gets the 2D array of u.

v_array()[source]: Gets the 2D array of v.

compute_distortion_matrix(method='raytrace')[source]

Computes the distortion matrix.

Parameters:: method (str, optional) – If “poly”, uses the pre-computed polynomial fit. The default (recommended) is “raytrace”.
Returns:: The 2x2 Jacobian matrix, d(xan,yan)/d(fpax,fpay); units of mm^-1.
Return type:: np.ndarray of float

compute_determinant()[source]

Determinant of distortion matrix.

Returns:: The determinant of the FPA –> field angle mapping, in mm^-2.
Return type:: float

load_pupil_mask(use_ray_trace=True)[source]

Loads the pupil mask.

Also sets the pupil coordinates.

Parameters:: use_ray_trace (bool, optional) – Ray traced pupil mask (Only turn off for testing, if stpsf-data is available.)
Returns:: The pupil mask.
Return type:: np.ndarray of float

path_diff(use_ray_trace=True)[source]

Path difference map. The behavior depends on the selected cycle.

The method is:

in Cycle 9, looks up the Zernike modes
in Cycle 10, uses the perturbation model

Returns:: The path difference map; same shape as self.u_array(). Units of microns.
Return type:: np.ndarray of float