"""
Copyright (c) 2010-2018 CNRS / Centre de Recherche Astrophysique de Lyon
Copyright (c) 2012-2016 Laure Piqueras <laure.piqueras@univ-lyon1.fr>
Copyright (c) 2014-2018 Simon Conseil <simon.conseil@univ-lyon1.fr>
Copyright (c) 2016 Martin Shepherd <martin.shepherd@univ-lyon1.fr>
All rights reserved.
Redistribution and use in source and binary forms, with or without
modification, are permitted provided that the following conditions are met:
1. Redistributions of source code must retain the above copyright notice, this
list of conditions and the following disclaimer.
2. Redistributions in binary form must reproduce the above copyright notice,
this list of conditions and the following disclaimer in the documentation
and/or other materials provided with the distribution.
3. Neither the name of the copyright holder nor the names of its contributors
may be used to endorse or promote products derived from this software
without specific prior written permission.
THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS" AND
ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE IMPLIED
WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE
DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT HOLDER OR CONTRIBUTORS BE LIABLE
FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL
DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR
SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER
CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY,
OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE
OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.
"""
import numbers
import numpy as np
from astropy.constants import c
from astropy.units import Quantity
__all__ = ('is_float', 'is_int', 'is_number', 'flux2mag', 'mag2flux',
'UnitArray', 'UnitMaskedArray', 'bounding_box')
[docs]
def is_float(x):
"""Test if `x` is a float number."""
return isinstance(x, numbers.Real)
[docs]
def is_int(x):
"""Test if `x` is an int number."""
return isinstance(x, numbers.Integral)
[docs]
def is_number(x):
"""Test if `x` is a number."""
return isinstance(x, numbers.Number)
[docs]
def flux2mag(flux, err_flux, wave):
"""Convert flux from erg.s-1.cm-2.A-1 to AB mag.
wave is the wavelength in A
"""
if flux > 0:
cs = c.to('Angstrom/s').value # speed of light in A/s
mag = -48.60 - 2.5 * np.log10(wave ** 2 * flux / cs)
err_mag = np.abs(2.5 * err_flux / (flux * np.log(10)))
return (mag, err_mag)
else:
return (99, np.inf)
[docs]
def mag2flux(mag, wave):
"""Convert flux from AB mag to erg.s-1.cm-2.A-1
wave is the wavelength in A
"""
cs = c.to('Angstrom/s').value # speed of light in A/s
return 10 ** (-0.4 * (mag + 48.60)) * cs / wave ** 2
[docs]
def bounding_box(form, center, radii, shape, posangle=0.0, step=None):
"""Return Y-axis and X-axis slice objects that bound a rectangular
image region that just encloses either an ellipse or a rectangle,
where the rectangle has a specified center position, Y-axis and X-axis
radii, and a given rotation angle relative to the image axes. The
effective center of the region is also returned.
If the ellipse or rectangle is partly outside of the image array,
the returned slices are clipped at the edges of the array. The
effective center that is returned, is the center before clipping.
Parameters
----------
form : str
The type of region whose rectangle image bounds are needed,
chosen from:
rectangle, a rotated rectangle or square.
ellipse, a rotated ellipse or circle.
center : float, float
The floating point array indexes of the centre of the circle,
in the order, y,x.
radii : float or float,float
The half-width and half-height of the ellipse or
rectangle. When the posangle is zero, the width and height are
parallel to the X and Y axes of the image array, respectively.
More generally, the width and height are along directions that
are posangle degrees counterclockwise of the X axis and Y axis,
respectively. If only one number is specified, then the width
and the height are both given that value.
posangle : float
The counterclockwise rotation angle of the chosen shape, in
degrees. When posangle is 0 degrees, the width and height of
the shape are along the X and Y axes of the image. Non-zero
values of posangle rotate the chosen shape anti-clockwise
of that position.
shape : int, int
The dimensions of the image array.
step : float, float
The per-pixel world-coordinate increments along the Y and X axes
of the image array, or [1.0,1.0] if radii is in pixels.
Returns
-------
out : clipped, unclipped, center
The 'clipped' return value is a list of the Y-axis and
X-axis slices to use to select all pixels within the
rectangular region of the image that just encloses the part of
the ellipse or rectangle that is within the image area.
The 'unclipped' return-value is the version of 'clipped' before
it was clipped at the edges of the image.
The 'center' return value is the pixel index of the center of
the region, prior to clipping.
If the region is entirely outside the range of an axis, a
zero-pixel slice is returned for that axis in 'clipped', with a
start value that is 0 if the pixels were all below pixel 0, or
shape-1 if they were all off the upper end of the range.
"""
# If only one radius is specified, use it as both the half-width and
# the half-height.
if np.isscalar(radii):
rx, ry = radii, radii
else:
rx, ry = radii
center = np.asarray(center)
if step is None:
step = (1.0, 1.0)
step = np.asarray(step)
# Convert the position angle to radians and precompute the sine and
# cosine of this.
pa = np.radians(posangle)
sin_pa = np.sin(pa)
cos_pa = np.cos(pa)
# Get the bounding box of a rotated rectangle?
if form == "rectangle":
# Calculate the maximum of the X-axis and Y-axis distances of the
# corners of the rotated rectangle from the rectangle's center.
xmax = abs(rx * cos_pa) + abs(ry * sin_pa)
ymax = abs(rx * sin_pa) + abs(ry * cos_pa)
# Get the bounding box of a rotated ellipse?
elif form == "ellipse":
# We use the following parametric equations for an unrotated
# ellipse with x and y along the image-array X and Y axes, where t
# is a parametric angle with no physical equivalent:
# x(pa=0) = rx * cos(t)
# y(pa=0) = ry * sin(t)
# We then rotate this anti-clockwise by pa:
# x(pa) = |cos(pa), -sin(pa)| |rx * cos(t)|
# y(pa) |sin(pa), cos(pa)| |ry * cos(t)|
# By differentiating the resulting equations of x(pa) and y(pa) by
# dt and setting the derivatives to zero, we obtain the following
# values for the angle t at which x(pa) and y(pa) are maximized.
t_xmax = np.arctan2(-ry * sin_pa, rx * cos_pa)
t_ymax = np.arctan2(ry * cos_pa, rx * sin_pa)
# Compute the half-width and half-height of the rectangle that
# encloses the ellipse, by computing the X and Y values,
# respectively, of the ellipse at the above angles.
xmax = np.abs(rx * np.cos(t_xmax) * cos_pa -
ry * np.sin(t_xmax) * sin_pa)
ymax = np.abs(rx * np.cos(t_ymax) * sin_pa +
ry * np.sin(t_ymax) * cos_pa)
else:
raise ValueError("form should be 'rectangle' or 'ellipse'")
# Put the height and width in an array, divide them by
# the pixel sizes along the Y and X axes to convert them to pixel
# counts, then convert them to the nearest integers.
w = np.floor(np.abs(np.array([2 * ymax, 2 * xmax]) / step) + 0.5).astype(int)
# Are the members of w even numbers of pixels?
iseven = np.mod(w, 2) == 0
# For each axis calculate the pixel index of the central pixel of
# the region where w is odd, or the index of the first of the two
# central pixels when w is even.
c = np.where(iseven, np.floor(center), np.floor(center + 0.5)).astype(int)
# Determine the indexes of the first and last pixels of the region
# along each axis, using integer arithmetic to avoid surprises.
first = np.where(iseven, c - w // 2 + 1, c - (w - 1) // 2)
last = np.where(iseven, c + w // 2, c + (w - 1) // 2)
# Calculate the effective center of the bounded region.
center = (first + last) / 2.0
# Compute the ideal slices that would select the bounding box.
ideal_slices = [slice(first[0], last[0] + 1), slice(first[1], last[1] + 1)]
# Are the selected pixels of the axes outside the bounds of the array?
max_indexes = np.asarray(shape) - 1
outside = np.logical_or(last < 0, first > max_indexes)
# Clip the first and last pixels to ensure that they lie within
# the bounds of the image.
imin, jmin = np.clip(first, (0, 0), max_indexes)
imax, jmax = np.clip(last, (0, 0), max_indexes)
# Compute the corresponding slices, replacing the clipped
# values by a zero-pixel range where the pre-clipped indexes
# were entirely outside the array.
clipped_slices = [slice(imin, (imax + 1) if not outside[0] else imax),
slice(jmin, (jmax + 1) if not outside[1] else jmax)]
# Return the ranges as slice objects, along with the effective
# center of the region.
return clipped_slices, ideal_slices, center
[docs]
def UnitArray(array, old_unit, new_unit):
if new_unit == old_unit:
return array
else:
return Quantity(array, old_unit, copy=False).to(new_unit).value
[docs]
def UnitMaskedArray(mask_array, old_unit, new_unit):
if new_unit == old_unit:
return mask_array
else:
return np.ma.array(Quantity(mask_array.data, old_unit, copy=False)
.to(new_unit).value, mask=mask_array.mask)