prysm.convolution¶
Defines behavior of convolvable items and a base class to encapsulate that behavior.
-
class
prysm.convolution.
Convolvable
(x, y, data, has_analytic_ft=False, labels=None, xy_unit=None, z_unit=None)¶ Bases:
prysm._richdata.RichData
A base class for convolvable objects to inherit from.
-
support_x
¶ Width of the domain in X.
-
support_y
¶ Width of the domain in Y.
-
support
¶ Width of the domain.
-
conv
(other)¶ Convolves this convolvable with another.
- Parameters
other (Convolvable) – A convolvable object
- Returns
a convolvable object
- Return type
Convolvable
Notes
If self and other both have analytic Fourier transforms, no math will be done and the aFTs are merged directly.
If only one of self or other has an analytic Fourier transform, the output grid will be defined by the object which does not have an analytic Fourier transform.
If neither has an analytic transform, the output grid will: - span max(self.support, other.support) - have sample spacing min(self.sample_spacing, other.sample_spacing)
This ensures the signal remains Nyquist sampled and (probably) doesn’t expand beyond the extent of the output window. The latter condition will be violated when two large convolvables are convolved.
-
deconv
(other, balance=1000, reg=None, is_real=True, clip=False, postnormalize=True)¶ Perform the deconvolution of this convolvable object by another.
- Parameters
other (Convolvable) – another convolvable object, used as the PSF in a Wiener deconvolution
balance (float, optional) – regularization parameter; passed through to skimage
reg (numpy.ndarray, optional) – regularization operator, passed through to skimage
is_real (bool, optional) – True if self and other are both real
clip (bool, optional) – clips self and other into (0,1)
postnormalize (bool, optional) – normalize the result such that it falls in [0,1]
- Returns
a new Convolable object
- Return type
Convolvable
Notes
See skimage: http://scikit-image.org/docs/dev/api/skimage.restoration.html#skimage.restoration.wiener
-
renorm
()¶ Renormalize so that the peak is at a value of unity and the minimum value is zero.
-
msaa
(factor=2)¶ Multi-Sample anti-aliasing.
Perform anti-aliasing by averaging blocks of (factor, factor) pixels into a simple value.
- Parameters
factor (int, optional) – factor by which to decimate the data
- Returns
self
- Return type
Convolvable
-
save
(path, nbits=8)¶ Write the image to a png, jpg, tiff, etc.
- Parameters
path (string) – path to write the image to
nbits (int) – number of bits in the output image
-
static
from_file
(path, scale)¶ Read a monochrome 8 bit per pixel file into a new Image instance.
- Parameters
path (string) – path to a file
scale (float) – pixel scale, in microns
- Returns
a new image object
- Return type
Convolvable
-
astype
(other_type)¶ Change this instance of one type into another.
Useful to access methods of the other class.
- Parameters
other_type (object) – the name of the other type to “cast” to, e.g. Interferogram. Not a string.
- Returns
type-converted to the other type.
- Return type
self
-
center_x
¶ Center “pixel” in x.
-
center_y
¶ Center “pixel” in y.
-
change_xy_unit
(to, inplace=True)¶ Change the x/y unit to a new one, scaling the data in the process.
- Parameters
to (astropy.unit or str) – if not an astropy unit, a string that is a valid attribute of astropy.units.
inplace (bool, optional) – if True, returns self. Otherwise returns the modified data.
- Returns
RichData – self, if inplace=True
numpy.ndarray, numpy.ndarray – x, y from self, if inplace=False
-
change_z_unit
(to, inplace=True)¶ Change the z unit to a new one, scaling the data in the process.
- Parameters
to (astropy.unit or str) – if not an astropy unit, a string that is a valid attribute of astropy.units.
inplace (bool, optional) – if True, returns self. Otherwise returns the modified data.
- Returns
RichData – self, if inplace=True
numpy.ndarray – data from self, if inplace=False
-
copy
()¶ Return a (deep) copy of this instance.
-
exact_polar
(rho, phi=None)¶ Retrieve data at the specified radial coordinates pairs.
- Parameters
rho (iterable) – radial coordinate(s) to sample
phi (iterable) – azimuthal coordinate(s) to sample
- Returns
data at the given points
- Return type
numpy.ndarray
-
exact_x
(x)¶ Return data at an exact x coordinate along the y=0 axis.
- Parameters
x (number or numpy.ndarray) – x coordinate(s) to return
- Returns
ndarray of values
- Return type
numpy.ndarray
-
exact_xy
(x, y=None)¶ Retrieve data at the specified X-Y frequency pairs.
- Parameters
x (iterable) – X coordinate(s) to retrieve
y (iterable) – Y coordinate(s) to retrieve
- Returns
data at the given points
- Return type
numpy.ndarray
-
exact_y
(y)¶ Return data at an exact y coordinate along the x=0 axis.
- Parameters
y (number or numpy.ndarray) – y coordinate(s) to return
- Returns
ndarray of values
- Return type
numpy.ndarray
-
plot2d
(xlim=None, ylim=None, clim=None, cmap=None, log=False, power=1, interpolation=None, show_colorbar=True, show_axlabels=True, fig=None, ax=None)¶ Plot the data in 2D.
- Parameters
xlim (float or iterable, optional) – x axis limits. If not iterable, symmetric version of the single value
ylim (float or iterable, optional) – y axis limits. If None and xlim is not None, copied from xlim. If not iterable, symmetric version of the single value.
clim (iterable, optional) – clim passed directly to matplotlib. If None, looked up on self._default_clim.
cmap (str, optional) – colormap to use, passed directly to matplotlib if not None. If None, looks up the default cmap for self._data_type on config
log (bool, optional) – if True, plot on a log color scale
power (float, optional) – if not 1, plot on a power stretched color scale
interpolation (str, optional) – interpolation method to use, passed directly to matplotlib
show_colorbar (bool, optional) – if True, draws the colorbar
show_axlabels (bool, optional) – if True, draws the axis labels
fig (matplotlib.figure.Figure) – Figure containing the plot
ax (matplotlib.axes.Axis) – Axis containing the plot
- Returns
fig (matplotlib.figure.Figure) – Figure containing the plot
ax (matplotlib.axes.Axis) – Axis containing the plot
-
sample_spacing
¶ center-to-center sample spacing.
-
samples_x
¶ Number of samples in the x dimension.
-
samples_y
¶ Number of samples in the y dimension.
-
shape
¶ Proxy to phase or data shape.
-
size
¶ Proxy to phase or data size.
-
slices
(twosided=None)¶ Create a Slices instance from this instance.
- Parameters
twosided (bool, optional) – if None, copied from self._default_twosided
- Returns
a Slices object
- Return type
Slices
-
-
class
prysm.convolution.
ConvolutionEngine
(c1, c2=None, spatial_finalization=(<built-in function abs>, ), Q=2, pad_method='linear_ramp')¶ Bases:
object
An engine to facilitate fine-grained control over convolutions.
-
fire
()¶ Convolve self.c1 and self.c2 with no fuss.
-
compute_kspace_data
()¶ Compute the k-space representation of the convolution of c1 and c2.
-
compute_kspace_units
()¶ Compute the k-space domain of the convolution of c1 and c2.
-
compute_spatial_units
()¶ Compute the spatial domain units of the convolution of c1 and c2.
-
ifft
()¶ Take the iFT to compute the spatial representation of the convolution of c1 and c2.
-
crop_output
()¶ Crop the output in the spatial domain to remove the padded area.
-
postprocess_spatial
()¶ Post-process the spatial domain.
-
merge_analytics
()¶ Merge c1 and c2 if they both have analytic FTs, else raise.
- Raises
ValueError – c1 or c2 does not have an analytic FT.
-
spatial
¶ Spatial representation, x, y, data.
-
kspace
¶ k-space representation, fx, fy, data.
-