def test_px_mm_strategy_equality():
simple = SimplePxMmStrategy()
assert simple == SimplePxMmStrategy()
parallax_corrected = ParallaxCorrectedPxMmStrategy(1, 1)
assert parallax_corrected == ParallaxCorrectedPxMmStrategy(1, 1)
assert parallax_corrected != ParallaxCorrectedPxMmStrategy(1, 2)
assert parallax_corrected != ParallaxCorrectedPxMmStrategy(2, 1)
assert parallax_corrected != simple
dx = flex.double(flex.grid(10, 10), 1)
dy = flex.double(flex.grid(10, 10), 1)
offset = OffsetPxMmStrategy(dx, dy)
assert offset == OffsetPxMmStrategy(dx, dy)
assert offset != simple
assert offset != parallax_corrected
offset_parallax_corrected = OffsetParallaxCorrectedPxMmStrategy(
1, 1, dx, dy)
assert offset_parallax_corrected == OffsetParallaxCorrectedPxMmStrategy(
1, 1, dx, dy)
assert offset_parallax_corrected != OffsetParallaxCorrectedPxMmStrategy(
1, 2, dx, dy)
assert offset_parallax_corrected != OffsetParallaxCorrectedPxMmStrategy(
2, 1, dx, dy)
assert offset_parallax_corrected != simple
assert offset_parallax_corrected != parallax_corrected
assert offset_parallax_corrected != offset
def test_panel_equality():
panel = create_detector(offset=0)[0]
panel2 = copy.deepcopy(panel)
assert panel == panel2
panel2.set_px_mm_strategy(ParallaxCorrectedPxMmStrategy(1, 1))
assert panel != panel2
panel.set_px_mm_strategy(ParallaxCorrectedPxMmStrategy(1, 1))
assert panel == panel2
def test_correction_on_random_coordinates(model):
convert = ParallaxCorrectedPxMmStrategy(model["mu"], model["t0"])
for i in range(10000):
xy = matrix.col((random.uniform(-1000, 1000), random.uniform(-1000, 1000)))
# Do the forward and reverse corrections
xy_corr_gold = matrix.col(correct_gold(model, xy))
xy_corr = matrix.col(convert.to_pixel(model["detector"], xy))
xy_corr_inv = matrix.col(convert.to_millimeter(model["detector"], xy_corr))
# Check the values
assert abs(xy_corr_gold - xy_corr) < 1e-7
assert abs(xy_corr_inv - xy) < 1e-3
def _detector(self):
"""Partly dummy detector"""
from scitbx import matrix
# Get the detector geometry
entry = self._h5_handle["entry"]
instrument = entry["instrument"]
detector = instrument["detector"]
# Initialise detector frame - origin at 0,0,0
fast = matrix.col((1.0, 0.0, 0.0))
slow = matrix.col((0.0, 1.0, 0.0))
orig = matrix.col((0.0, 0.0, 0.0))
# Get the pixel and image size
pixel_size = (
1.0e-3 * detector["x_pixel_size"].value,
1.0e-3 * detector["y_pixel_size"].value,
)
layout = detector["layout"].value[0].split("X")
image_size = int(layout[0]), int(layout[1])
trusted_range = (-1, detector["saturation_value"][0])
thickness = float(detector["sensor_thickness"].value) / 1000.0
material = str(detector["sensor_material"].value[0])
# Make the detector
detector = self._detector_factory.make_detector(
"PAD",
fast,
slow,
orig,
pixel_size,
image_size,
trusted_range,
name="Panel",
thickness=thickness,
material=material,
)
# At the moment, beam is a dummy object because wavelength is not set in
# the header. Therefore, the px<-->mm strategy will generally be
# incorrect. Set it anyway, to override later.
beam = self._beam()
wavelength = beam.get_wavelength()
from cctbx.eltbx import attenuation_coefficient
from dxtbx.model import ParallaxCorrectedPxMmStrategy
# this will fail for undefined composite materials
table = attenuation_coefficient.get_table(material)
# mu_at_angstrom returns cm^-1, but need mu in mm^-1
mu = table.mu_at_angstrom(wavelength) / 10.0
for panel in detector:
panel.set_mu(mu)
panel.set_px_mm_strategy(
ParallaxCorrectedPxMmStrategy(mu, thickness))
return detector
def _detector(self):
"""Return a working detector instance, with added mask regions."""
detector = self._detector_factory.imgCIF_H(self._get_cbf_handle(), "PAD")
for f0, f1, s0, s1 in determine_pilatus_mask(detector):
detector[0].add_mask(f0 - 1, s0 - 1, f1, s1)
m = re.search(
r"^#\s*(\S+)\ssensor, thickness\s*([0-9.]+)\s*m\s*$",
self._cif_header,
re.MULTILINE,
)
if m:
# header gives thickness in metres, we store mm
thickness = float(m.group(2)) * 1000
material = m.group(1)
if material == "Silicon":
material = "Si"
for panel in detector:
panel.set_thickness(thickness)
panel.set_material(material)
try:
# a header only CBF file will not have a beam object
beam = self._beam()
except Exception:
beam = None
if beam:
# attenuation coefficient depends on the beam wavelength
wavelength = beam.get_wavelength()
# this will fail for undefined composite materials
table = attenuation_coefficient.get_table(material)
# mu_at_angstrom returns cm^-1
mu = table.mu_at_angstrom(wavelength) / 10.0
for panel in detector:
panel.set_px_mm_strategy(
ParallaxCorrectedPxMmStrategy(mu, thickness)
)
panel.set_mu(mu)
m = re.search(r"^#\s*Detector:\s+(.*?)\s*$", self._cif_header, re.MULTILINE)
if m and m.group(1):
panel.set_identifier(m.group(1).encode())
size = detector[0].get_image_size()
if size == (2463, 2527):
self.vendortype = "Pilatus-6M"
elif size == (1475, 1679):
self.vendortype = "Pilatus-2M"
elif size == (487, 619):
self.vendortype = "Pilatus-300K"
return detector
def get_values(invert_y):
from dxtbx.model.beam import beam_factory
beam = beam_factory.simple(wavelength=1)
if invert_y:
y_direction = "-y"
else:
y_direction = "+y"
from dxtbx.model.detector import detector_factory
detector = detector_factory.simple(
sensor=detector_factory.sensor("PAD"),
distance=100,
beam_centre=[50, 50],
fast_direction="+x",
slow_direction=y_direction,
pixel_size=[0.1, 0.1],
image_size=[1000, 1000])
from dxtbx.model import ParallaxCorrectedPxMmStrategy
from cctbx.eltbx import attenuation_coefficient
wavelength = beam.get_wavelength()
thickness = 0.5
table = attenuation_coefficient.get_table("Si")
mu = table.mu_at_angstrom(wavelength) / 10.0
t0 = thickness
for panel in detector:
panel.set_px_mm_strategy(ParallaxCorrectedPxMmStrategy(mu, t0))
v1 = detector[0].pixel_to_millimeter((0, 0))
v2 = detector[0].pixel_to_millimeter((1000, 1000))
return v1, v2
def test_parallax_correction():
# Attenuation length
table = attenuation_coefficient.get_table("Si")
mu = table.mu_at_angstrom(1) / 10.0
t0 = 0.320
# Create the detector
detector = Detector(
Panel(
"", # Type
"", # Name
(10, 0, 0), # Fast axis
(0, 10, 0), # Slow axis
(0, 0, 200), # Origin
(0.172, 0.172), # Pixel size
(512, 512), # Image size
(0, 1000), # Trusted range
0.0, # Thickness
"", # Material
ParallaxCorrectedPxMmStrategy(mu, t0),
)
)
for i in range(10000):
mm = (random.uniform(-1000, 1000), random.uniform(-1000, 1000))
px = detector[0].millimeter_to_pixel(mm)
mm2 = detector[0].pixel_to_millimeter(px)
assert abs(matrix.col(mm) - matrix.col(mm2)) < 1e-3
def _detector(self):
d = FormatCBFMultiTileHierarchyStill._detector(self)
try:
# a header only CBF file will not have a beam object
beam = self._beam()
except Exception as e:
if "CBF_NOTFOUND" not in str(e):
raise e
return d
# take into consideration here the thickness of the sensor also the
# wavelength of the radiation (which we have in the same file...)
wavelength = beam.get_wavelength()
thickness = 0.5 # mm, see Hart et al. 2012
from cctbx.eltbx import attenuation_coefficient
table = attenuation_coefficient.get_table("Si")
# mu_at_angstrom returns cm^-1
mu = table.mu_at_angstrom(wavelength) / 10.0 # mu: mm^-1
t0 = thickness
for panel in d:
panel.set_px_mm_strategy(ParallaxCorrectedPxMmStrategy(mu, t0))
return d
def _detector(self):
'''Return a model for a simple detector, presuming no one has
one of these on a two-theta stage. Assert that the beam centre is
provided in the Mosflm coordinate frame.'''
distance = float(
self._cif_header_dictionary['Detector_distance'].split()[0])
beam_xy = self._cif_header_dictionary['Beam_xy'].replace(
'(', '').replace(')', '').replace(',', '').split()[:2]
wavelength = float(
self._cif_header_dictionary['Wavelength'].split()[0])
beam_x, beam_y = map(float, beam_xy)
pixel_xy = self._cif_header_dictionary['Pixel_size'].replace(
'm', '').replace('x', '').split()
pixel_x, pixel_y = map(float, pixel_xy)
if 'Silicon' in self._cif_header_dictionary:
thickness = float(
self._cif_header_dictionary['Silicon'].split()[2]) * 1000.0
material = 'Si'
elif 'CdTe' in self._cif_header_dictionary:
thickness = float(
self._cif_header_dictionary['CdTe'].split()[2]) * 1000.0
material = 'CdTe'
else:
thickness = 0.450
material = 'Si'
nx = int(
self._cif_header_dictionary['X-Binary-Size-Fastest-Dimension'])
ny = int(self._cif_header_dictionary['X-Binary-Size-Second-Dimension'])
overload = dxtbx_overload_scale * int(
self._cif_header_dictionary['Count_cutoff'].split()[0])
underload = -1
# take into consideration here the thickness of the sensor also the
# wavelength of the radiation (which we have in the same file...)
from cctbx.eltbx import attenuation_coefficient
table = attenuation_coefficient.get_table(material)
mu = table.mu_at_angstrom(wavelength) / 10.0
t0 = thickness
detector = self._detector_factory.simple(
'PAD', distance * 1000.0,
(beam_x * pixel_x * 1000.0, beam_y * pixel_y * 1000.0), '+x', '-y',
(1000 * pixel_x, 1000 * pixel_y), (nx, ny), (underload, overload),
[], ParallaxCorrectedPxMmStrategy(mu, t0))
detector[0].set_thickness(thickness)
detector[0].set_material(material)
detector[0].set_mu(mu)
return detector
def _detector(self):
""" Create an Eiger detector profile (taken from FormatCBFMiniEiger) """
configuration = self.header["configuration"]
info = self.header["info"]
distance = configuration["detector_distance"]
wavelength = configuration["wavelength"]
beam_x = configuration["beam_center_x"]
beam_y = configuration["beam_center_y"]
pixel_x = configuration["x_pixel_size"]
pixel_y = configuration["y_pixel_size"]
material = configuration["sensor_material"]
thickness = configuration["sensor_thickness"] * 1000
nx = configuration["x_pixels_in_detector"]
ny = configuration["y_pixels_in_detector"]
if "count_rate_correction_count_cutoff" in configuration:
overload = configuration["count_rate_correction_count_cutoff"]
else:
# hard-code if missing from Eiger stream header
overload = 4001400
underload = -1
try:
identifier = configuration["description"]
except KeyError:
identifier = "Unknown Eiger"
table = attenuation_coefficient.get_table(material)
mu = table.mu_at_angstrom(wavelength) / 10.0
t0 = thickness
detector = self._detector_factory.simple(
sensor="PAD",
distance=distance * 1000.0,
beam_centre=(beam_x * pixel_x * 1000.0, beam_y * pixel_y * 1000.0),
fast_direction="+x",
slow_direction="-y",
pixel_size=(1000 * pixel_x, 1000 * pixel_y),
image_size=(nx, ny),
trusted_range=(underload, overload),
mask=[],
px_mm=ParallaxCorrectedPxMmStrategy(mu, t0),
mu=mu,
)
for f0, f1, s0, s1 in determine_eiger_mask(detector):
detector[0].add_mask(f0 - 1, s0 - 1, f1, s1)
for panel in detector:
panel.set_thickness(thickness)
panel.set_material(material)
panel.set_identifier(identifier)
panel.set_mu(mu)
return detector
def _detector(self):
"""Return a model for a simple detector, presuming no one has
one of these on a two-theta stage. Assert that the beam centre is
provided in the Mosflm coordinate frame."""
distance = float(
self._cif_header_dictionary["Detector_distance"].split()[0])
beam_xy = (self._cif_header_dictionary["Beam_xy"].replace(
"(", "").replace(")", "").replace(",", "").split()[:2])
wavelength = float(
self._cif_header_dictionary["Wavelength"].split()[0])
beam_x, beam_y = map(float, beam_xy)
pixel_xy = (self._cif_header_dictionary["Pixel_size"].replace(
"m", "").replace("x", "").split())
pixel_x, pixel_y = map(float, pixel_xy)
thickness = float(
self._cif_header_dictionary["Silicon"].split()[2]) * 1000.0
nx = int(
self._cif_header_dictionary["X-Binary-Size-Fastest-Dimension"])
ny = int(self._cif_header_dictionary["X-Binary-Size-Second-Dimension"])
overload = int(self._cif_header_dictionary["Count_cutoff"].split()[0])
underload = -1
# take into consideration here the thickness of the sensor also the
# wavelength of the radiation (which we have in the same file...)
from cctbx.eltbx import attenuation_coefficient
table = attenuation_coefficient.get_table("Si")
mu = table.mu_at_angstrom(wavelength) / 10.0
t0 = thickness
# FIXME would also be very nice to be able to take into account the
# misalignment of the individual modules given the calibration information...
detector = self._detector_factory.simple(
"PAD",
distance * 1000.0,
(beam_x * pixel_x * 1000.0, beam_y * pixel_y * 1000.0),
"+x",
"-y",
(1000 * pixel_x, 1000 * pixel_y),
(nx, ny),
(underload, overload),
[],
ParallaxCorrectedPxMmStrategy(mu, t0),
)
"""
for f0, f1, s0, s1 in determine_pilatus_mask(detector):
detector[0].add_mask(f0-1, s0-1, f1, s1)
"""
return detector
def _detector(self, index=None):
from dxtbx.model.detector import Detector
from scitbx import matrix
import math
wavelength = self.get_beam(index).get_wavelength()
from dxtbx.model import ParallaxCorrectedPxMmStrategy
from cctbx.eltbx import attenuation_coefficient
table = attenuation_coefficient.get_table("Si")
mu = table.mu_at_angstrom(wavelength) / 10.0
px_mm = ParallaxCorrectedPxMmStrategy(mu, self.thickness)
if self.RECONST_MODE:
return self._detector_factory.simple(
sensor="PAD",
distance=self.distance,
beam_centre=(
self.RECONST_SIZE / 2 * self.PIXEL_SIZE,
self.RECONST_SIZE / 2 * self.PIXEL_SIZE,
),
fast_direction="-x",
slow_direction="-y",
pixel_size=(self.PIXEL_SIZE, self.PIXEL_SIZE),
image_size=(self.RECONST_SIZE, self.RECONST_SIZE),
trusted_range=(-1, 65535),
mask=[],
) # TODO: add gaps
detector = Detector()
root = detector.hierarchy()
root.set_frame((-1, 0, 0), (0, 1, 0), (0, 0, -self.distance))
for i in range(8):
angle = math.pi * self.panel_rotations[i] / 180.0
fast = matrix.col((math.cos(angle), math.sin(angle), 0))
slow = matrix.col((-math.sin(angle), math.cos(angle), 0))
normal = fast.cross(slow)
origin = (matrix.col((
-self.panel_origins[i][0],
self.panel_origins[i][1],
self.panel_origins[i][2],
)) / 1000.0)
p = root.add_panel()
p.set_type("SENSOR_PAD")
p.set_name("Panel%d" % i)
p.set_image_size((512, 1024))
p.set_trusted_range((-1, 65535))
p.set_pixel_size((self.PIXEL_SIZE, self.PIXEL_SIZE))
p.set_thickness(self.thickness)
p.set_local_frame(fast.elems, slow.elems, origin.elems)
p.set_px_mm_strategy(px_mm)
p.set_gain(10)
return detector
def _detector(self):
'''Return a working detector instance, with added mask regions.'''
detector = self._detector_factory.imgCIF_H(self._get_cbf_handle(),
'PAD')
for f0, s0, f1, s1 in determine_pilatus_mask(detector):
detector[0].add_mask(f0, s0, f1, s1)
import re
m = re.search('^#\s*(\S+)\ssensor, thickness\s*([0-9.]+)\s*m\s*$', \
self._cif_header, re.MULTILINE)
if m:
# header gives thickness in metres, we store mm
thickness = float(m.group(2)) * 1000
material = m.group(1)
if material == 'Silicon':
material = 'Si'
for panel in detector:
panel.set_thickness(thickness)
panel.set_material(material)
try:
# a header only CBF file will not have a beam object
beam = self._beam()
except Exception:
beam = None
if beam:
# attenuation coefficient depends on the beam wavelength
wavelength = beam.get_wavelength()
from cctbx.eltbx import attenuation_coefficient
from dxtbx.model import ParallaxCorrectedPxMmStrategy
# this will fail for undefined composite materials
table = attenuation_coefficient.get_table(material)
# mu_at_angstrom returns cm^-1
mu = table.mu_at_angstrom(wavelength) / 10.0
for panel in detector:
panel.set_px_mm_strategy(
ParallaxCorrectedPxMmStrategy(mu, thickness))
panel.set_mu(mu)
m = re.search('^#\s*Detector:\s+(.*?)\s*$', \
self._cif_header, re.MULTILINE)
if m and m.group(1):
panel.set_identifier(m.group(1))
return detector
def _detector(self):
distance = float(
self._cif_header_dictionary['Detector_distance'].split()[0])
beam_xy = self._cif_header_dictionary['Beam_xy'].replace(
'(', '').replace(')', '').replace(',', '').split()[:2]
wavelength = float(
self._cif_header_dictionary['Wavelength'].split()[0])
beam_x, beam_y = map(float, beam_xy)
pixel_xy = self._cif_header_dictionary['Pixel_size'].replace(
'm', '').replace('x', '').split()
pixel_x, pixel_y = map(float, pixel_xy)
thickness = float(
self._cif_header_dictionary['Silicon'].split()[-2]) * 1000.0
nx = int(
self._cif_header_dictionary['X-Binary-Size-Fastest-Dimension'])
ny = int(
self._cif_header_dictionary['X-Binary-Size-Second-Dimension'])
overload = int(
self._cif_header_dictionary['Count_cutoff'].split()[0])
underload = -1
from cctbx.eltbx import attenuation_coefficient
table = attenuation_coefficient.get_table("Si")
mu = table.mu_at_angstrom(wavelength) / 10.0
t0 = thickness
detector = self._detector_factory.simple(
'PAD', distance * 1000.0, (beam_x * pixel_x * 1000.0,
beam_y * pixel_y * 1000.0), '+x', '-y',
(1000 * pixel_x, 1000 * pixel_y),
(nx, ny), (underload, overload), [],
ParallaxCorrectedPxMmStrategy(mu, t0))
for f0, s0, f1, s1 in [(0, 513, 1030, 550)]:
detector[0].add_mask(f0, s0, f1, s1)
return detector
class FormatCBFCspad(FormatCBFMultiTileHierarchy, FormatStill):
'''An image reading class CSPAD CBF files'''
@staticmethod
def understand(image_file):
'''Check to see if this looks like an CSPD CBF format image, i.e. we can
make sense of it.'''
cbf_handle = pycbf.cbf_handle_struct()
cbf_handle.read_widefile(image_file, pycbf.MSG_DIGEST)
cbf_handle.find_category("diffrn_detector")
if cbf_handle.count_rows() > 1:
return False # support 1 detector per file for now
cbf_handle.find_column("type")
return cbf_handle.get_value() == "CS PAD"
def _detector(self):
d = FormatCBFMultiTileHierarchy._detector(self)
try:
# a header only CBF file will not have a beam object
beam = self._beam()
except Exception, e:
if "CBF_NOTFOUND" not in str(e): raise e
return d
# take into consideration here the thickness of the sensor also the
# wavelength of the radiation (which we have in the same file...)
wavelength = beam.get_wavelength()
thickness = 0.5
from cctbx.eltbx import attenuation_coefficient
table = attenuation_coefficient.get_table("Si")
# mu_at_angstrom returns cm^-1
mu = table.mu_at_angstrom(wavelength) / 10.0
t0 = thickness
for panel in d:
panel.set_px_mm_strategy(ParallaxCorrectedPxMmStrategy(mu, t0))
return d
def _detector(self):
"""Return a model for the detector, allowing for two-theta offsets
and the detector position. This will be rather more complex..."""
detector_name = self._header_dictionary["DETECTOR_NAMES"].split()[0].strip()
detector_axes = self.get_detector_axes(detector_name)
fast = matrix.col(tuple(detector_axes[:3]))
slow = matrix.col(tuple(detector_axes[3:]))
distortion = self.get_distortion(detector_name)
# multiply through by the distortion to get the true detector fast, slow
detector_fast, detector_slow = (
distortion[0] * fast + distortion[1] * slow,
distortion[2] * fast + distortion[3] * slow,
)
beam_pixels = self.get_beam_pixels(detector_name)
pixel_size = self.get_pixel_size(detector_name)
image_size = self.get_image_size(detector_name)
detector_origin = -(
beam_pixels[0] * pixel_size[0] * detector_fast
+ beam_pixels[1] * pixel_size[1] * detector_slow
)
gonio_axes = self.get_gonio_axes(detector_name)
gonio_values = self.get_gonio_values(detector_name)
gonio_units = self._header_dictionary["%sGONIO_UNITS" % detector_name].split()
gonio_num_axes = int(
self._header_dictionary["%sGONIO_NUM_VALUES" % detector_name]
)
rotations = []
translations = []
for j, unit in enumerate(gonio_units):
axis = matrix.col(gonio_axes[3 * j : 3 * (j + 1)])
if unit == "deg":
rotations.append(
axis.axis_and_angle_as_r3_rotation_matrix(gonio_values[j], deg=True)
)
translations.append(matrix.col((0.0, 0.0, 0.0)))
elif unit == "mm":
rotations.append(
matrix.sqr((1.0, 0.0, 0.0, 0.0, 1.0, 0.0, 0.0, 0.0, 1.0))
)
translations.append(gonio_values[j] * axis)
else:
raise RuntimeError("unknown axis unit %s" % unit)
rotations.reverse()
translations.reverse()
for j in range(gonio_num_axes):
detector_fast = rotations[j] * detector_fast
detector_slow = rotations[j] * detector_slow
detector_origin = rotations[j] * detector_origin
detector_origin = translations[j] + detector_origin
overload = int(float(self._header_dictionary["SATURATED_VALUE"]))
underload = -1
# Unfortunately thickness and material are not stored in the header. Set
# to sensible defaults.
t0 = 0.450
material = "Si"
table = attenuation_coefficient.get_table(material)
wavelength = float(self._header_dictionary["SCAN_WAVELENGTH"])
mu = table.mu_at_angstrom(wavelength) / 10.0
detector = self._detector_factory.complex(
"PAD",
detector_origin.elems,
detector_fast.elems,
detector_slow.elems,
pixel_size,
image_size,
(underload, overload),
px_mm=ParallaxCorrectedPxMmStrategy(mu, t0),
mu=mu,
)
for panel in detector:
panel.set_thickness(t0)
panel.set_material(material)
return detector
from __future__ import division
import sys
from dxtbx.serialize import load
from dxtbx.model import ParallaxCorrectedPxMmStrategy, SimplePxMmStrategy
from scitbx.array_family import flex
from matplotlib import pylab
convert = ParallaxCorrectedPxMmStrategy(0.252500934883)
convert2 = SimplePxMmStrategy()
sweep = load.imageset(sys.argv[1])
detector = sweep.get_detector()
#print detector[0].pixel_to_millimeter((0, 1))
image_size = sweep[0].all()
image = flex.double(flex.grid(image_size))
for j in range(image_size[0]):
for i in range(image_size[1]):
mm1 = convert2.to_millimeter(detector[0], (i, j))
px2 = convert.to_pixel(detector[0], mm1)
image[j,i] = j - px2[1]
print flex.max(image), flex.min(image)
pylab.imshow(image.as_numpy_array())
pylab.show()
def _detector(self):
"""Return a model for a simple detector, presuming no one has
one of these on a two-theta stage. Assert that the beam centre is
provided in the Mosflm coordinate frame."""
if not self._multi_panel:
detector = FormatCBFMini._detector(self)
for f0, f1, s0, s1 in determine_pilatus_mask(detector):
detector[0].add_mask(f0 - 1, s0 - 1, f1, s1)
return detector
# got to here means 60-panel version
d = Detector()
distance = float(
self._cif_header_dictionary["Detector_distance"].split()[0])
beam_xy = (self._cif_header_dictionary["Beam_xy"].replace(
"(", "").replace(")", "").replace(",", "").split()[:2])
beam_x, beam_y = map(float, beam_xy)
wavelength = float(
self._cif_header_dictionary["Wavelength"].split()[0])
pixel_xy = (self._cif_header_dictionary["Pixel_size"].replace(
"m", "").replace("x", "").split())
pixel_x, pixel_y = map(float, pixel_xy)
thickness = float(
self._cif_header_dictionary["Silicon"].split()[2]) * 1000.0
nx = int(
self._cif_header_dictionary["X-Binary-Size-Fastest-Dimension"])
ny = int(self._cif_header_dictionary["X-Binary-Size-Second-Dimension"])
overload = int(self._cif_header_dictionary["Count_cutoff"].split()[0])
underload = -1
# take into consideration here the thickness of the sensor also the
# wavelength of the radiation (which we have in the same file...)
table = attenuation_coefficient.get_table("Si")
mu = table.mu_at_angstrom(wavelength) / 10.0
t0 = thickness
# FIXME would also be very nice to be able to take into account the
# misalignment of the individual modules given the calibration...
# single detector or multi-module detector
pixel_x *= 1000.0
pixel_y *= 1000.0
distance *= 1000.0
beam_centre = matrix.col((beam_x * pixel_x, beam_y * pixel_y, 0))
fast = matrix.col((1.0, 0.0, 0.0))
slow = matrix.col((0.0, -1.0, 0.0))
s0 = matrix.col((0, 0, -1))
origin = (distance * s0) - (fast * beam_centre[0]) - (slow *
beam_centre[1])
root = d.hierarchy()
root.set_local_frame(fast.elems, slow.elems, origin.elems)
det = _DetectorDatabase["Pilatus"]
# Edge dead areas not included, only gaps between modules matter
n_fast, remainder = divmod(nx, det.module_size_fast)
assert (n_fast - 1) * det.gap_fast == remainder
n_slow, remainder = divmod(ny, det.module_size_slow)
assert (n_slow - 1) * det.gap_slow == remainder
mx = det.module_size_fast
my = det.module_size_slow
dx = det.gap_fast
dy = det.gap_slow
xmins = [(mx + dx) * i for i in range(n_fast)]
xmaxes = [mx + (mx + dx) * i for i in range(n_fast)]
ymins = [(my + dy) * i for i in range(n_slow)]
ymaxes = [my + (my + dy) * i for i in range(n_slow)]
self.coords = {}
fast = matrix.col((1.0, 0.0, 0.0))
slow = matrix.col((0.0, 1.0, 0.0))
panel_idx = 0
for ymin, ymax in zip(ymins, ymaxes):
for xmin, xmax in zip(xmins, xmaxes):
xmin_mm = xmin * pixel_x
ymin_mm = ymin * pixel_y
origin_panel = fast * xmin_mm + slow * ymin_mm
panel_name = "Panel%d" % panel_idx
panel_idx += 1
p = d.add_panel()
p.set_type("SENSOR_PAD")
p.set_name(panel_name)
p.set_raw_image_offset((xmin, ymin))
p.set_image_size((xmax - xmin, ymax - ymin))
p.set_trusted_range((underload, overload))
p.set_pixel_size((pixel_x, pixel_y))
p.set_thickness(thickness)
p.set_material("Si")
p.set_mu(mu)
p.set_px_mm_strategy(ParallaxCorrectedPxMmStrategy(mu, t0))
p.set_local_frame(fast.elems, slow.elems, origin_panel.elems)
p.set_raw_image_offset((xmin, ymin))
self.coords[panel_name] = (xmin, ymin, xmax, ymax)
return d
def _detector(self, index=None):
if index is None:
index = 0
run = self.get_run_from_index(index)
det = self._get_psana_detector(run)
geom = det.pyda.geoaccess(run.run())
cob = read_slac_metrology(geometry=geom, include_asic_offset=True)
distance = env_distance(
self.params.detector_address[0], run.env(), self.params.cspad.detz_offset
)
d = Detector()
pg0 = d.hierarchy()
# first deal with D0
det_num = 0
origin = col((cob[(0,)] * col((0, 0, 0, 1)))[0:3])
fast = col((cob[(0,)] * col((1, 0, 0, 1)))[0:3]) - origin
slow = col((cob[(0,)] * col((0, 1, 0, 1)))[0:3]) - origin
origin += col((0.0, 0.0, -distance))
pg0.set_local_frame(fast.elems, slow.elems, origin.elems)
pg0.set_name("D%d" % (det_num))
for quad_num in range(4):
# Now deal with Qx
pg1 = pg0.add_group()
origin = col((cob[(0, quad_num)] * col((0, 0, 0, 1)))[0:3])
fast = col((cob[(0, quad_num)] * col((1, 0, 0, 1)))[0:3]) - origin
slow = col((cob[(0, quad_num)] * col((0, 1, 0, 1)))[0:3]) - origin
pg1.set_local_frame(fast.elems, slow.elems, origin.elems)
pg1.set_name("D%dQ%d" % (det_num, quad_num))
for sensor_num in range(8):
# Now deal with Sy
pg2 = pg1.add_group()
origin = col((cob[(0, quad_num, sensor_num)] * col((0, 0, 0, 1)))[0:3])
fast = (
col((cob[(0, quad_num, sensor_num)] * col((1, 0, 0, 1)))[0:3])
- origin
)
slow = (
col((cob[(0, quad_num, sensor_num)] * col((0, 1, 0, 1)))[0:3])
- origin
)
pg2.set_local_frame(fast.elems, slow.elems, origin.elems)
pg2.set_name("D%dQ%dS%d" % (det_num, quad_num, sensor_num))
# Now deal with Az
for asic_num in range(2):
val = "ARRAY_D0Q%dS%dA%d" % (quad_num, sensor_num, asic_num)
p = pg2.add_panel()
origin = col(
(cob[(0, quad_num, sensor_num, asic_num)] * col((0, 0, 0, 1)))[
0:3
]
)
fast = (
col(
(
cob[(0, quad_num, sensor_num, asic_num)]
* col((1, 0, 0, 1))
)[0:3]
)
- origin
)
slow = (
col(
(
cob[(0, quad_num, sensor_num, asic_num)]
* col((0, 1, 0, 1))
)[0:3]
)
- origin
)
p.set_local_frame(fast.elems, slow.elems, origin.elems)
p.set_pixel_size(
(cspad_cbf_tbx.pixel_size, cspad_cbf_tbx.pixel_size)
)
p.set_image_size(cspad_cbf_tbx.asic_dimension)
p.set_trusted_range(
(
cspad_tbx.cspad_min_trusted_value,
cspad_tbx.cspad_saturated_value,
)
)
p.set_name(val)
try:
beam = self._beam(index)
except Exception:
print(
"No beam object initialized. Returning CSPAD detector without parallax corrections"
)
return d
# take into consideration here the thickness of the sensor also the
# wavelength of the radiation (which we have in the same file...)
wavelength = beam.get_wavelength()
thickness = 0.5 # mm, see Hart et al. 2012
table = attenuation_coefficient.get_table("Si")
# mu_at_angstrom returns cm^-1
mu = table.mu_at_angstrom(wavelength) / 10.0 # mu: mm^-1
t0 = thickness
for panel in d:
panel.set_px_mm_strategy(ParallaxCorrectedPxMmStrategy(mu, t0))
return d
def test_detector():
from dxtbx.model import ParallaxCorrectedPxMmStrategy
def create_detector(offset=0):
# Create the detector
detector = Detector(
Panel(
"", # Type
"Panel", # Name
(10, 0, 0), # Fast axis
(0, 10, 0), # Slow axis
(0 + offset, 0 + offset, 200 - offset),
# Origin
(0.172, 0.172), # Pixel size
(512, 512), # Image size
(0, 1000), # Trusted range
0.1, # Thickness
"Si", # Material
identifier="123")) # Identifier
return detector
detector = create_detector()
# Perform some tests
tst_get_identifier(detector)
tst_get_gain(detector)
tst_set_mosflm_beam_centre(detector)
tst_get_pixel_lab_coord(detector)
tst_get_image_size_mm(detector)
tst_is_value_in_trusted_range(detector)
tst_is_coord_valid(detector)
tst_pixel_to_millimeter_to_pixel(detector)
tst_get_names(detector)
tst_get_thickness(detector)
tst_get_material(detector)
tst_resolution(detector)
tst_panel_mask()
# Attenuation length
from cctbx.eltbx import attenuation_coefficient
table = attenuation_coefficient.get_table("Si")
mu = table.mu_at_angstrom(1) / 10.0
t0 = 0.320
# Create another detector with different origin
detector_moved = create_detector(offset=100)
tst_detectors_are_different(detector, detector_moved)
detector_moved_copy = create_detector(offset=100)
tst_detectors_are_same(detector_moved, detector_moved_copy)
# Create the detector
detector = Detector(
Panel(
"", # Type
"", # Name
(10, 0, 0), # Fast axis
(0, 10, 0), # Slow axis
(0, 0, 200), # Origin
(0.172, 0.172), # Pixel size
(512, 512), # Image size
(0, 1000), # Trusted range
0.0, # Thickness
"", # Material
ParallaxCorrectedPxMmStrategy(mu, t0)))
tst_parallax_correction(detector)
def _detector(self):
'''Detector model, allowing for small offsets in the positions of 60
detector modules'''
# Module positional offsets in x, y, in pixels - for the moment ignoring the
# rotational offsets as these are not well defined. To be honest these
# positional offsets are also not well defined as I do not know how they
# should be applied...
x = {
(0, 0): -0.477546,
(0, 1): 0.130578,
(0, 2): 0.045041,
(0, 3): -0.439872,
(0, 4): -0.382077,
(1, 0): 0.087405,
(1, 1): 0.249597,
(1, 2): 0.184265,
(1, 3): 0.158342,
(1, 4): 0.025225,
(2, 0): -0.179892,
(2, 1): -0.010974,
(2, 2): -0.139207,
(2, 3): 0.282851,
(2, 4): -0.442219,
(3, 0): -0.185027,
(3, 1): 0.218601,
(3, 2): 0.092585,
(3, 3): 0.35862,
(3, 4): -0.29161,
(4, 0): 0.145368,
(4, 1): 0.609289,
(4, 2): 0.396265,
(4, 3): 0.41625,
(4, 4): 0.07152,
(5, 0): 0.247142,
(5, 1): 0.046563,
(5, 2): 0.248714,
(5, 3): -0.044628,
(5, 4): -0.391509,
(6, 0): 0.516643,
(6, 1): 0.358453,
(6, 2): 0.069219,
(6, 3): 0.095861,
(6, 4): -0.167403,
(7, 0): -0.381352,
(7, 1): -0.35338,
(7, 2): 0.348656,
(7, 3): 0.024543,
(7, 4): 0.328706,
(8, 0): 0.150886,
(8, 1): 0.244987,
(8, 2): -0.102911,
(8, 3): 0.16633,
(8, 4): 0.386622,
(9, 0): 0.037924,
(9, 1): 0.314392,
(9, 2): 0.238818,
(9, 3): 0.815028,
(9, 4): -0.048818,
(10, 0): -0.670524,
(10, 1): -0.304119,
(10, 2): 0.252284,
(10, 3): -0.05485,
(10, 4): -0.355264,
(11, 0): -0.404947,
(11, 1): -0.020622,
(11, 2): 0.648473,
(11, 3): -0.277175,
(11, 4): -0.711951
}
y = {
(0, 0): -0.494797,
(0, 1): -0.212976,
(0, 2): 0.085351,
(0, 3): 0.35494,
(0, 4): 0.571189,
(1, 0): -0.421708,
(1, 1): 0.061914,
(1, 2): 0.238996,
(1, 3): 0.146692,
(1, 4): 0.407145,
(2, 0): -0.313212,
(2, 1): -0.225025,
(2, 2): 0.031613,
(2, 3): -0.047839,
(2, 4): 0.42716,
(3, 0): -0.361193,
(3, 1): 0.057663,
(3, 2): 0.022357,
(3, 3): 0.062717,
(3, 4): 0.150611,
(4, 0): 0.035511,
(4, 1): -0.271567,
(4, 2): 0.007761,
(4, 3): -0.124021,
(4, 4): 0.093017,
(5, 0): -0.238897,
(5, 1): -0.179724,
(5, 2): -0.113608,
(5, 3): 0.017841,
(5, 4): -0.012933,
(6, 0): -0.166337,
(6, 1): -0.272922,
(6, 2): -0.194665,
(6, 3): -0.058535,
(6, 4): -0.405404,
(7, 0): -0.318824,
(7, 1): -0.311276,
(7, 2): -0.205223,
(7, 3): -0.292664,
(7, 4): -0.474762,
(8, 0): -0.039504,
(8, 1): -0.239887,
(8, 2): -0.343485,
(8, 3): -0.459429,
(8, 4): -0.426901,
(9, 0): -0.187805,
(9, 1): 0.282727,
(9, 2): -0.601164,
(9, 3): -0.467605,
(9, 4): -0.589271,
(10, 0): 0.028311,
(10, 1): -0.391571,
(10, 2): -0.463112,
(10, 3): -0.358092,
(10, 4): -0.285396,
(11, 0): 0.01863,
(11, 1): -0.380099,
(11, 2): -0.234953,
(11, 3): -0.593992,
(11, 4): -0.801247
}
distance = float(
self._cif_header_dictionary['Detector_distance'].split()[0])
beam_xy = self._cif_header_dictionary['Beam_xy'].replace(
'(', '').replace(')', '').replace(',', '').split()[:2]
beam_x, beam_y = map(float, beam_xy)
wavelength = float(
self._cif_header_dictionary['Wavelength'].split()[0])
pixel_xy = self._cif_header_dictionary['Pixel_size'].replace(
'm', '').replace('x', '').split()
pixel_x, pixel_y = map(float, pixel_xy)
thickness = float(
self._cif_header_dictionary['Silicon'].split()[2]) * 1000.0
nx = int(
self._cif_header_dictionary['X-Binary-Size-Fastest-Dimension'])
ny = int(self._cif_header_dictionary['X-Binary-Size-Second-Dimension'])
overload = int(self._cif_header_dictionary['Count_cutoff'].split()[0])
underload = -1
# take into consideration here the thickness of the sensor also the
# wavelength of the radiation (which we have in the same file...)
from cctbx.eltbx import attenuation_coefficient
table = attenuation_coefficient.get_table("Si")
mu = table.mu_at_angstrom(wavelength) / 10.0
t0 = thickness
# FIXME would also be very nice to be able to take into account the
# misalignment of the individual modules given the calibration...
# single detector or multi-module detector
pixel_x *= 1000.0
pixel_y *= 1000.0
distance *= 1000.0
if not self._multi_panel:
detector = self._detector_factory.simple(
'PAD', distance, (beam_x * pixel_x, beam_y * pixel_y), '+x',
'-y', (pixel_x, pixel_y), (nx, ny), (underload, overload), [],
ParallaxCorrectedPxMmStrategy(mu, t0))
for f0, f1, s0, s1 in determine_pilatus_mask(detector):
detector[0].add_mask(f0 - 1, s0 - 1, f1, s1)
detector[0].set_thickness(thickness)
detector[0].set_material('Si')
detector[0].set_mu(mu)
return detector
# got to here means 60-panel version
from dxtbx.model import Detector
from scitbx import matrix
d = Detector()
beam_centre = matrix.col((beam_x * pixel_x, beam_y * pixel_y, 0))
fast = matrix.col((1.0, 0.0, 0.0))
slow = matrix.col((0.0, -1.0, 0.0))
s0 = matrix.col((0, 0, -1))
origin = (distance * s0) - (fast * beam_centre[0]) - \
(slow * beam_centre[1])
root = d.hierarchy()
root.set_local_frame(fast.elems, slow.elems, origin.elems)
xmins = [0, 494, 988, 1482, 1976]
xmaxes = [487, 981, 1475, 1969, 2463]
ymins = [
0, 212, 424, 636, 848, 1060, 1272, 1484, 1696, 1908, 2120, 2332
]
ymaxes = [
195, 407, 619, 831, 1043, 1255, 1467, 1679, 1891, 2103, 2315, 2527
]
self.coords = {}
fast = matrix.col((1.0, 0.0, 0.0))
slow = matrix.col((0.0, 1.0, 0.0))
panel_idx = 0
for ymin, ymax in zip(ymins, ymaxes):
for xmin, xmax in zip(xmins, xmaxes):
xmin_mm = xmin * pixel_x
ymin_mm = ymin * pixel_y
origin_panel = fast * xmin_mm + slow * ymin_mm
panel_name = "Panel%d" % panel_idx
panel_idx += 1
p = root.add_panel()
p.set_type("SENSOR_PAD")
p.set_px_mm_strategy(ParallaxCorrectedPxMmStrategy(mu, t0))
p.set_name(panel_name)
p.set_image_size((xmax - xmin, ymax - ymin))
p.set_trusted_range((underload, overload))
p.set_pixel_size((pixel_x, pixel_y))
p.set_thickness(thickness)
p.set_material('Si')
p.set_mu(mu)
p.set_local_frame(fast.elems, slow.elems, origin_panel.elems)
p.set_raw_image_offset((xmin, ymin))
self.coords[panel_name] = (xmin, ymin, xmax, ymax)
return d
def correct_inv(self, xy):
from dxtbx.model import ParallaxCorrectedPxMmStrategy
convert = ParallaxCorrectedPxMmStrategy(self.mu, self.t0)
return convert.to_millimeter(self.detector[0], xy)
def _detector(self):
# module positions from detector blueprints - modelling at the moment as
# 24 modules, each consisting of 5 sensors (the latter is ignored)
from dxtbx.model import Detector
from scitbx import matrix
import math
x = matrix.col((-1, 0, 0))
y = matrix.col((0, 1, 0))
z = matrix.col((0, 0, 1))
beam_xy = self._cif_header_dictionary['Beam_xy']
beam_xy = beam_xy.replace('(', '').replace(')',
'').replace(',',
'').split()[:2]
obs_beam_x, obs_beam_y = [float(f) for f in beam_xy]
ideal_beam_x = 1075
ideal_beam_y = 2594
beam_shift_x = 0.172 * (ideal_beam_x - obs_beam_x)
beam_shift_y = 0.172 * (ideal_beam_y - obs_beam_y)
distance = float(self._cif_header_dictionary['Detector_distance'].
split()[0]) * 1000.0
wavelength = float(
self._cif_header_dictionary['Wavelength'].split()[0])
thickness = float(
self._cif_header_dictionary['Silicon'].split()[2]) * 1000.0
off_x = 184.9
detector = Detector()
root = detector.hierarchy()
root.set_frame(x.elems, y.elems, (-distance * z + (beam_shift_x * x) +
(beam_shift_y * y)).elems)
from cctbx.eltbx import attenuation_coefficient
table = attenuation_coefficient.get_table("Si")
mu = table.mu_at_angstrom(wavelength) / 10.0
t0 = thickness
px_mm = ParallaxCorrectedPxMmStrategy(mu, t0)
self.coords = {}
for j in range(24):
shift_y = 195 + 17
ymin, ymax = j * shift_y, j * shift_y + 195
angle = math.pi * (-12.2 + 0.5 * 7.903 + j *
(7.903 + 0.441)) / 180.0
fast = matrix.col((1, 0, 0))
slow = matrix.col((0, math.sin(angle), math.cos(angle)))
normal = fast.cross(slow)
row_origin = 250.0 * normal - off_x * fast - 16.8 * slow
if not self._multi_panel:
xmin, xmax = 0, 2463
# OK two calls to add_panel here for detector like things => two
# copies of the panel then? https://github.com/dials/dials/issues/189
# ... this is also not the source of the leak
# OBS! you need to set the panel to a root before set local frame...
p = root.add_panel()
p.set_type('SENSOR_PAD')
p.set_name('row-%02d' % j)
p.set_raw_image_offset((xmin, ymin))
p.set_image_size((2463, 195))
p.set_trusted_range((-1, 1000000))
p.set_pixel_size((0.172, 0.172))
p.set_local_frame(fast.elems, slow.elems, row_origin.elems)
p.set_thickness(thickness)
p.set_material('Si')
p.set_mu(mu)
p.set_px_mm_strategy(px_mm)
p.set_raw_image_offset((xmin, ymin))
self.coords[p.get_name()] = (xmin, ymin, xmax, ymax)
else:
shift_x = 487 + 7
for i in range(5):
xmin, xmax = i * shift_x, i * shift_x + 487
origin = row_origin + i * (487 + 7) * 0.172 * fast
# OBS! you need to set the panel to a root before set local frame...
p = root.add_panel()
p.set_type('SENSOR_PAD')
p.set_name('row-%02d-col-%02d' % (j, i))
p.set_raw_image_offset((xmin, ymin))
p.set_image_size((487, 195))
p.set_trusted_range((-1, 1000000))
p.set_pixel_size((0.172, 0.172))
p.set_local_frame(fast.elems, slow.elems, origin.elems)
p.set_thickness(thickness)
p.set_material('Si')
p.set_mu(mu)
p.set_px_mm_strategy(px_mm)
p.set_raw_image_offset((xmin, ymin))
self.coords[p.get_name()] = (xmin, ymin, xmax, ymax)
return detector
def _detector(self):
"""Return a model for a simple detector, presuming no one has
one of these on a two-theta stage. Assert that the beam centre is
provided in the Mosflm coordinate frame."""
distance = float(self._cif_header_dictionary["Detector_distance"].split()[0])
beam_xy = (
self._cif_header_dictionary["Beam_xy"]
.replace("(", "")
.replace(")", "")
.replace(",", "")
.split()[:2]
)
wavelength = float(self._cif_header_dictionary["Wavelength"].split()[0])
beam_x, beam_y = map(float, beam_xy)
pixel_xy = (
self._cif_header_dictionary["Pixel_size"]
.replace("m", "")
.replace("x", "")
.split()
)
pixel_x, pixel_y = map(float, pixel_xy)
if "Silicon" in self._cif_header_dictionary:
thickness = (
float(self._cif_header_dictionary["Silicon"].split()[2]) * 1000.0
)
material = "Si"
sensor = "PAD"
elif "CdTe" in self._cif_header_dictionary:
thickness = float(self._cif_header_dictionary["CdTe"].split()[2]) * 1000.0
material = "CdTe"
sensor = "PAD"
elif "CCD" in self._cif_header_dictionary:
thickness = 0
material = None
sensor = "CCD"
else:
thickness = 0
material = None
sensor = None
nx = int(self._cif_header_dictionary["X-Binary-Size-Fastest-Dimension"])
ny = int(self._cif_header_dictionary["X-Binary-Size-Second-Dimension"])
overload = dxtbx_overload_scale * int(
self._cif_header_dictionary["Count_cutoff"].split()[0]
)
underload = -1
if material is not None:
# take into consideration here the thickness of the sensor also the
# wavelength of the radiation (which we have in the same file...)
from cctbx.eltbx import attenuation_coefficient
table = attenuation_coefficient.get_table(material)
mu = table.mu_at_angstrom(wavelength) / 10.0
t0 = thickness
px_mm = ParallaxCorrectedPxMmStrategy(mu, t0)
else:
px_mm = SimplePxMmStrategy()
detector = self._detector_factory.simple(
sensor,
distance * 1000.0,
(beam_x * pixel_x * 1000.0, beam_y * pixel_y * 1000.0),
"+x",
"-y",
(1000 * pixel_x, 1000 * pixel_y),
(nx, ny),
(underload, overload),
[],
px_mm=px_mm,
)
if material is not None:
detector[0].set_thickness(thickness)
detector[0].set_material(material)
detector[0].set_mu(mu)
return detector
def _detector(self):
'''Return a model for the detector, allowing for two-theta offsets
and the detector position. This will be rather more complex...'''
detector_name = self._header_dictionary['DETECTOR_NAMES'].split(
)[0].strip()
detector_axes = map(
float, self._header_dictionary['%sDETECTOR_VECTORS' %
detector_name].split())
fast = matrix.col(tuple(detector_axes[:3]))
slow = matrix.col(tuple(detector_axes[3:]))
distortion = map(
int, self._header_dictionary['%sSPATIAL_DISTORTION_VECTORS' %
detector_name].split())
# multiply through by the distortion to get the true detector fast, slow
detector_fast, detector_slow = distortion[0] * fast + distortion[1] * slow, \
distortion[2] * fast + distortion[3] * slow
beam_pixels = map(
float, self._header_dictionary['%sSPATIAL_DISTORTION_INFO' %
detector_name].split()[:2])
pixel_size = map(
float, self._header_dictionary['%sSPATIAL_DISTORTION_INFO' %
detector_name].split()[2:])
image_size = map(
int, self._header_dictionary['%sDETECTOR_DIMENSIONS' %
detector_name].split())
detector_origin = - (beam_pixels[0] * pixel_size[0] * detector_fast + \
beam_pixels[1] * pixel_size[1] * detector_slow)
gonio_axes = map(
float,
self._header_dictionary['%sGONIO_VECTORS' % detector_name].split())
gonio_values = map(
float,
self._header_dictionary['%sGONIO_VALUES' % detector_name].split())
gonio_units = self._header_dictionary['%sGONIO_UNITS' %
detector_name].split()
gonio_num_axes = int(self._header_dictionary['%sGONIO_NUM_VALUES' %
detector_name])
rotations = []
translations = []
for j, unit in enumerate(gonio_units):
axis = matrix.col(gonio_axes[3 * j:3 * (j + 1)])
if unit == 'deg':
rotations.append(
axis.axis_and_angle_as_r3_rotation_matrix(gonio_values[j],
deg=True))
translations.append(matrix.col((0.0, 0.0, 0.0)))
elif unit == 'mm':
rotations.append(
matrix.sqr((1.0, 0.0, 0.0, 0.0, 1.0, 0.0, 0.0, 0.0, 1.0)))
translations.append(gonio_values[j] * axis)
else:
raise RuntimeError('unknown axis unit %s' % unit)
rotations.reverse()
translations.reverse()
for j in range(gonio_num_axes):
detector_fast = rotations[j] * detector_fast
detector_slow = rotations[j] * detector_slow
detector_origin = rotations[j] * detector_origin
detector_origin = translations[j] + detector_origin
overload = int(float(self._header_dictionary['SATURATED_VALUE']))
underload = -1
# Unfortunately thickness and material are not stored in the header. Set
# to sensible defaults.
t0 = 0.450
material = 'Si'
from cctbx.eltbx import attenuation_coefficient
table = attenuation_coefficient.get_table(material)
wavelength = float(self._header_dictionary['SCAN_WAVELENGTH'])
mu = table.mu_at_angstrom(wavelength) / 10.0
detector = self._detector_factory.complex(
'PAD',
detector_origin.elems,
detector_fast.elems,
detector_slow.elems,
pixel_size,
image_size, (underload, overload),
px_mm=ParallaxCorrectedPxMmStrategy(mu, t0),
mu=mu)
for panel in detector:
panel.set_thickness(t0)
panel.set_material(material)
return detector
def _detector(self, index=None):
import psana
from xfel.cftbx.detector.cspad_cbf_tbx import read_slac_metrology
from dxtbx.model import Detector
from scitbx.matrix import col
from dxtbx.model import ParallaxCorrectedPxMmStrategy
if index is None: index = 0
self._env = self._ds.env()
self._det = psana.Detector(self._src, self._env)
geom = self._det.pyda.geoaccess(self.events_list[index])
cob = read_slac_metrology(geometry=geom, include_asic_offset=True)
d = Detector()
pg0 = d.hierarchy()
# first deal with D0
det_num = 0
origin = col((cob[(0, )] * col((0, 0, 0, 1)))[0:3])
fast = col((cob[(0, )] * col((1, 0, 0, 1)))[0:3]) - origin
slow = col((cob[(0, )] * col((0, 1, 0, 1)))[0:3]) - origin
origin += col((0., 0., -100.))
pg0.set_local_frame(fast.elems, slow.elems, origin.elems)
pg0.set_name('D%d' % (det_num))
for quad_num in xrange(4):
# Now deal with Qx
pg1 = pg0.add_group()
origin = col((cob[(0, quad_num)] * col((0, 0, 0, 1)))[0:3])
fast = col((cob[(0, quad_num)] * col((1, 0, 0, 1)))[0:3]) - origin
slow = col((cob[(0, quad_num)] * col((0, 1, 0, 1)))[0:3]) - origin
pg1.set_local_frame(fast.elems, slow.elems, origin.elems)
pg1.set_name('D%dQ%d' % (det_num, quad_num))
for sensor_num in xrange(8):
# Now deal with Sy
pg2 = pg1.add_group()
origin = col((cob[(0, quad_num, sensor_num)] * col(
(0, 0, 0, 1)))[0:3])
fast = col((cob[(0, quad_num, sensor_num)] * col(
(1, 0, 0, 1)))[0:3]) - origin
slow = col((cob[(0, quad_num, sensor_num)] * col(
(0, 1, 0, 1)))[0:3]) - origin
pg2.set_local_frame(fast.elems, slow.elems, origin.elems)
pg2.set_name('D%dQ%dS%d' % (det_num, quad_num, sensor_num))
# Now deal with Az
for asic_num in xrange(2):
val = 'ARRAY_D0Q%dS%dA%d' % (quad_num, sensor_num,
asic_num)
p = pg2.add_panel()
origin = col(
(cob[(0, quad_num, sensor_num, asic_num)] * col(
(0, 0, 0, 1)))[0:3])
fast = col((cob[(0, quad_num, sensor_num, asic_num)] * col(
(1, 0, 0, 1)))[0:3]) - origin
slow = col((cob[(0, quad_num, sensor_num, asic_num)] * col(
(0, 1, 0, 1)))[0:3]) - origin
p.set_local_frame(fast.elems, slow.elems, origin.elems)
p.set_pixel_size(
(cspad_cbf_tbx.pixel_size, cspad_cbf_tbx.pixel_size))
p.set_image_size(cspad_cbf_tbx.asic_dimension)
p.set_trusted_range((cspad_tbx.cspad_min_trusted_value,
cspad_tbx.cspad_saturated_value))
p.set_name(val)
try:
beam = self._beam(index)
except Exception:
print 'No beam object initialized. Returning CSPAD detector without parallax corrections'
return d
# take into consideration here the thickness of the sensor also the
# wavelength of the radiation (which we have in the same file...)
wavelength = beam.get_wavelength()
thickness = 0.5 # mm, see Hart et al. 2012
from cctbx.eltbx import attenuation_coefficient
table = attenuation_coefficient.get_table("Si")
# mu_at_angstrom returns cm^-1
mu = table.mu_at_angstrom(wavelength) / 10.0 # mu: mm^-1
t0 = thickness
for panel in d:
panel.set_px_mm_strategy(ParallaxCorrectedPxMmStrategy(mu, t0))
return d
def _detector(self, index=None):
from PSCalib.SegGeometryStore import sgs
from xfel.cftbx.detector.cspad_cbf_tbx import basis_from_geo
run = self.get_run_from_index(index)
if run.run() in self._cached_detector:
return self._cached_detector[run.run()]
if index is None:
index = 0
self._env = self._ds.env()
assert len(self.params.detector_address) == 1
self._det = psana.Detector(self.params.detector_address[0], self._env)
geom = self._det.pyda.geoaccess(self._get_event(index).run())
pixel_size = (self._det.pixel_size(self._get_event(index)) / 1000.0
) # convert to mm
d = Detector()
pg0 = d.hierarchy()
# first deal with D0
det_num = 0
root = geom.get_top_geo()
root_basis = basis_from_geo(root)
while len(root.get_list_of_children()) == 1:
sub = root.get_list_of_children()[0]
sub_basis = basis_from_geo(sub)
root = sub
root_basis = root_basis * sub_basis
t = root_basis.translation
root_basis.translation = col((t[0], t[1], -t[2]))
origin = col((root_basis * col((0, 0, 0, 1)))[0:3])
fast = col((root_basis * col((1, 0, 0, 1)))[0:3]) - origin
slow = col((root_basis * col((0, 1, 0, 1)))[0:3]) - origin
pg0.set_local_frame(fast.elems, slow.elems, origin.elems)
pg0.set_name("D%d" % (det_num))
# Now deal with modules
for module_num in range(len(root.get_list_of_children())):
module = root.get_list_of_children()[module_num]
module_basis = basis_from_geo(module)
origin = col((module_basis * col((0, 0, 0, 1)))[0:3])
fast = col((module_basis * col((1, 0, 0, 1)))[0:3]) - origin
slow = col((module_basis * col((0, 1, 0, 1)))[0:3]) - origin
pg1 = pg0.add_group()
pg1.set_local_frame(fast.elems, slow.elems, origin.elems)
pg1.set_name("D%dM%d" % (det_num, module_num))
# Read the known layout of the Jungfrau 2x4 module
sg = sgs.Create(segname=module.oname)
xx, yy = sg.get_seg_xy_maps_um()
xx = xx / 1000
yy = yy / 1000
# Now deal with ASICs
for asic_num in range(8):
val = "ARRAY_D0M%dA%d" % (module_num, asic_num)
dim_slow = xx.shape[0]
dim_fast = xx.shape[1]
sensor_id = asic_num // 4 # There are 2X4 asics per module
asic_in_sensor_id = asic_num % 4 # this number will be 0,1,2 or 3
id_slow = sensor_id * (dim_slow // 2)
id_fast = asic_in_sensor_id * (dim_fast // 4)
origin = col((xx[id_slow][id_fast], yy[id_slow][id_fast], 0))
fp = col(
(xx[id_slow][id_fast + 1], yy[id_slow][id_fast + 1], 0))
sp = col(
(xx[id_slow + 1][id_fast], yy[id_slow + 1][id_fast], 0))
fast = (fp - origin).normalize()
slow = (sp - origin).normalize()
p = pg1.add_panel()
p.set_local_frame(fast.elems, slow.elems, origin.elems)
p.set_pixel_size((pixel_size, pixel_size))
p.set_trusted_range((-10, 2e6))
p.set_name(val)
thickness, material = 0.32, "Si"
p.set_thickness(thickness) # mm
p.set_material(material)
# Compute the attenuation coefficient.
# This will fail for undefined composite materials
# mu_at_angstrom returns cm^-1, but need mu in mm^-1
table = attenuation_coefficient.get_table(material)
wavelength = self.get_beam(index).get_wavelength()
mu = table.mu_at_angstrom(wavelength) / 10.0
p.set_mu(mu)
p.set_px_mm_strategy(
ParallaxCorrectedPxMmStrategy(mu, thickness))
if (self.params.jungfrau.use_big_pixels
and os.environ.get("DONT_USE_BIG_PIXELS_JUNGFRAU") is
None):
p.set_image_size((1030, 514))
break
else:
p.set_image_size((dim_fast // 4, dim_slow // 2))
if (self.params.jungfrau.use_big_pixels
and os.environ.get("DONT_USE_BIG_PIXELS_JUNGFRAU") is None):
assert len(d) == 8
self._cached_detector[run.run()] = d
return d
def _detector(self):
distance = float(
self._cif_header_dictionary['Detector_distance'].split()[0])
beam_xy = self._cif_header_dictionary['Beam_xy'].replace(
'(', '').replace(')', '').replace(',', '').split()[:2]
wavelength = float(
self._cif_header_dictionary['Wavelength'].split()[0])
beam_x, beam_y = map(float, beam_xy)
pixel_xy = self._cif_header_dictionary['Pixel_size'].replace(
'm', '').replace('x', '').split()
pixel_x, pixel_y = map(float, pixel_xy)
# extract sensor thickness, assume < 10mm; default to 450 microns unfound
try:
thickness = float(
self._cif_header_dictionary['Silicon'].split()[-2]) * 1000.0
material = 'Si'
if thickness > 10:
thickness = thickness / 1000.0
except KeyError:
thickness = 0.450
material = 'Si'
nx = int(
self._cif_header_dictionary['X-Binary-Size-Fastest-Dimension'])
ny = int(self._cif_header_dictionary['X-Binary-Size-Second-Dimension'])
if 'Count_cutoff' in self._cif_header_dictionary:
overload = int(
self._cif_header_dictionary['Count_cutoff'].split()[0])
else:
# missing from data transformed with GPhL converter - dials#376
overload = 100000000
underload = -1
try:
identifier = self._cif_header_dictionary['Detector']
except KeyError:
identifier = 'Unknown Eiger'
from cctbx.eltbx import attenuation_coefficient
table = attenuation_coefficient.get_table("Si")
mu = table.mu_at_angstrom(wavelength) / 10.0
t0 = thickness
detector = self._detector_factory.simple(
'PAD', distance * 1000.0,
(beam_x * pixel_x * 1000.0, beam_y * pixel_y * 1000.0), '+x', '-y',
(1000 * pixel_x, 1000 * pixel_y), (nx, ny), (underload, overload),
[], ParallaxCorrectedPxMmStrategy(mu, t0))
for f0, f1, s0, s1 in determine_eiger_mask(detector):
detector[0].add_mask(f0 - 1, s0 - 1, f1, s1)
for panel in detector:
panel.set_thickness(thickness)
panel.set_material(material)
panel.set_identifier(identifier)
panel.set_mu(mu)
return detector
def _detector(self):
'''Return a working detector instance, with added mask regions.'''
cif_header = FormatCBFFullPilatus.get_cbf_header(self._image_file)
self._cif_header_dictionary = {}
for record in cif_header.split('\n'):
if not '#' in record[:1]:
continue
if len(record[1:].split()) <= 2 and record.count(':') == 2:
self._cif_header_dictionary['timestamp'] = record[1:].strip()
continue
tokens = record.replace('=', '').replace(':', '').split()[1:]
self._cif_header_dictionary[tokens[0]] = ' '.join(tokens[1:])
for record in self._mime_header.split('\n'):
if not record.strip():
continue
token, value = record.split(':')
self._cif_header_dictionary[token.strip()] = value.strip()
distance = float(
self._cif_header_dictionary['Detector_distance'].split()[0])
beam_xy = self._cif_header_dictionary['Beam_xy'].replace(
'(', '').replace(')', '').replace(',', '').split()[:2]
wavelength = float(
self._cif_header_dictionary['Wavelength'].split()[0])
beam_x, beam_y = map(float, beam_xy)
pixel_xy = self._cif_header_dictionary['Pixel_size'].replace(
'm', '').replace('x', '').split()
pixel_x, pixel_y = map(float, pixel_xy)
nx = int(
self._cif_header_dictionary['X-Binary-Size-Fastest-Dimension'])
ny = int(self._cif_header_dictionary['X-Binary-Size-Second-Dimension'])
overload = int(self._cif_header_dictionary['Count_cutoff'].split()[0])
underload = -1
two_theta = float(
self._cif_header_dictionary['Detector_2theta'].split()[0])
# hard code "correct" values for these
distance = 0.11338
beam_x = 222.0
beam_y = 383.0
detector = self._detector_factory.two_theta(
'PAD', distance * 1000.0,
(beam_x * pixel_x * 1000.0, beam_y * pixel_y * 1000.0), '+x', '-y',
'+x', two_theta, (1000 * pixel_x, 1000 * pixel_y), (nx, ny),
(underload, overload), [], None)
import re
m = re.search('^#\s*(\S+)\ssensor, thickness\s*([0-9.]+)\s*m\s*$', \
self._cif_header, re.MULTILINE)
if m:
# header gives thickness in metres, we store mm
thickness = float(m.group(2)) * 1000
material = m.group(1)
if material == 'Silicon':
material = 'Si'
for panel in detector:
panel.set_thickness(thickness)
panel.set_material(material)
try:
# a header only CBF file will not have a beam object
beam = self._beam()
except Exception:
pass
if beam:
# attenuation coefficient depends on the beam wavelength
wavelength = beam.get_wavelength()
from cctbx.eltbx import attenuation_coefficient
from dxtbx.model import ParallaxCorrectedPxMmStrategy
# this will fail for undefined composite materials
table = attenuation_coefficient.get_table(material)
# mu_at_angstrom returns cm^-1, but need mu in mm^-1
mu = table.mu_at_angstrom(wavelength) / 10.0
for panel in detector:
panel.set_mu(mu)
panel.set_px_mm_strategy(
ParallaxCorrectedPxMmStrategy(mu, thickness))
return detector
def _detector(self):
'''Detector model, allowing for small offsets in the positions of 60
detector modules'''
# Module positional offsets in x, y, in pixels - for the moment ignoring the
# rotational offsets as these are not well defined. To be honest these
# positional offsets are also not well defined as I do not know how they
# should be applied...
x = {
(0, 0): -0.477546,
(0, 1): 0.130578,
(0, 2): 0.045041,
(0, 3): -0.439872,
(0, 4): -0.382077,
(1, 0): 0.087405,
(1, 1): 0.249597,
(1, 2): 0.184265,
(1, 3): 0.158342,
(1, 4): 0.025225,
(2, 0): -0.179892,
(2, 1): -0.010974,
(2, 2): -0.139207,
(2, 3): 0.282851,
(2, 4): -0.442219,
(3, 0): -0.185027,
(3, 1): 0.218601,
(3, 2): 0.092585,
(3, 3): 0.35862,
(3, 4): -0.29161,
(4, 0): 0.145368,
(4, 1): 0.609289,
(4, 2): 0.396265,
(4, 3): 0.41625,
(4, 4): 0.07152,
(5, 0): 0.247142,
(5, 1): 0.046563,
(5, 2): 0.248714,
(5, 3): -0.044628,
(5, 4): -0.391509,
(6, 0): 0.516643,
(6, 1): 0.358453,
(6, 2): 0.069219,
(6, 3): 0.095861,
(6, 4): -0.167403,
(7, 0): -0.381352,
(7, 1): -0.35338,
(7, 2): 0.348656,
(7, 3): 0.024543,
(7, 4): 0.328706,
(8, 0): 0.150886,
(8, 1): 0.244987,
(8, 2): -0.102911,
(8, 3): 0.16633,
(8, 4): 0.386622,
(9, 0): 0.037924,
(9, 1): 0.314392,
(9, 2): 0.238818,
(9, 3): 0.815028,
(9, 4): -0.048818,
(10, 0): -0.670524,
(10, 1): -0.304119,
(10, 2): 0.252284,
(10, 3): -0.05485,
(10, 4): -0.355264,
(11, 0): -0.404947,
(11, 1): -0.020622,
(11, 2): 0.648473,
(11, 3): -0.277175,
(11, 4): -0.711951
}
y = {
(0, 0): -0.494797,
(0, 1): -0.212976,
(0, 2): 0.085351,
(0, 3): 0.35494,
(0, 4): 0.571189,
(1, 0): -0.421708,
(1, 1): 0.061914,
(1, 2): 0.238996,
(1, 3): 0.146692,
(1, 4): 0.407145,
(2, 0): -0.313212,
(2, 1): -0.225025,
(2, 2): 0.031613,
(2, 3): -0.047839,
(2, 4): 0.42716,
(3, 0): -0.361193,
(3, 1): 0.057663,
(3, 2): 0.022357,
(3, 3): 0.062717,
(3, 4): 0.150611,
(4, 0): 0.035511,
(4, 1): -0.271567,
(4, 2): 0.007761,
(4, 3): -0.124021,
(4, 4): 0.093017,
(5, 0): -0.238897,
(5, 1): -0.179724,
(5, 2): -0.113608,
(5, 3): 0.017841,
(5, 4): -0.012933,
(6, 0): -0.166337,
(6, 1): -0.272922,
(6, 2): -0.194665,
(6, 3): -0.058535,
(6, 4): -0.405404,
(7, 0): -0.318824,
(7, 1): -0.311276,
(7, 2): -0.205223,
(7, 3): -0.292664,
(7, 4): -0.474762,
(8, 0): -0.039504,
(8, 1): -0.239887,
(8, 2): -0.343485,
(8, 3): -0.459429,
(8, 4): -0.426901,
(9, 0): -0.187805,
(9, 1): 0.282727,
(9, 2): -0.601164,
(9, 3): -0.467605,
(9, 4): -0.589271,
(10, 0): 0.028311,
(10, 1): -0.391571,
(10, 2): -0.463112,
(10, 3): -0.358092,
(10, 4): -0.285396,
(11, 0): 0.01863,
(11, 1): -0.380099,
(11, 2): -0.234953,
(11, 3): -0.593992,
(11, 4): -0.801247
}
distance = float(
self._cif_header_dictionary['Detector_distance'].split()[0])
beam_xy = self._cif_header_dictionary['Beam_xy'].replace(
'(', '').replace(')', '').replace(',', '').split()[:2]
beam_x, beam_y = map(float, beam_xy)
wavelength = float(
self._cif_header_dictionary['Wavelength'].split()[0])
pixel_xy = self._cif_header_dictionary['Pixel_size'].replace(
'm', '').replace('x', '').split()
pixel_x, pixel_y = map(float, pixel_xy)
thickness = float(
self._cif_header_dictionary['Silicon'].split()[2]) * 1000.0
nx = int(
self._cif_header_dictionary['X-Binary-Size-Fastest-Dimension'])
ny = int(self._cif_header_dictionary['X-Binary-Size-Second-Dimension'])
overload = int(self._cif_header_dictionary['Count_cutoff'].split()[0])
underload = -1
# take into consideration here the thickness of the sensor also the
# wavelength of the radiation (which we have in the same file...)
from cctbx.eltbx import attenuation_coefficient
table = attenuation_coefficient.get_table("Si")
mu = table.mu_at_angstrom(wavelength) / 10.0
t0 = thickness
# FIXME would also be very nice to be able to take into account the
# misalignment of the individual modules given the calibration information...
detector = self._detector_factory.simple(
'PAD', distance * 1000.0,
(beam_x * pixel_x * 1000.0, beam_y * pixel_y * 1000.0), '+x', '-y',
(1000 * pixel_x, 1000 * pixel_y), (nx, ny), (underload, overload),
[], ParallaxCorrectedPxMmStrategy(mu, t0))
for f0, f1, s0, s1 in determine_pilatus_mask(detector):
detector[0].add_mask(f0 - 1, s0 - 1, f1, s1)
return detector
def _detector(self):
distance = float(self._cif_header_dictionary["Detector_distance"].split()[0])
beam_xy = (
self._cif_header_dictionary["Beam_xy"]
.replace("(", "")
.replace(")", "")
.replace(",", "")
.split()[:2]
)
wavelength = float(self._cif_header_dictionary["Wavelength"].split()[0])
beam_x, beam_y = map(float, beam_xy)
pixel_xy = (
self._cif_header_dictionary["Pixel_size"]
.replace("m", "")
.replace("x", "")
.split()
)
pixel_x, pixel_y = map(float, pixel_xy)
if "Silicon" in self._cif_header_dictionary:
thickness = (
float(self._cif_header_dictionary["Silicon"].split()[2]) * 1000.0
)
material = "Si"
elif "CdTe" in self._cif_header_dictionary:
thickness = float(self._cif_header_dictionary["CdTe"].split()[2]) * 1000.0
material = "CdTe"
else:
thickness = 0.450
material = "Si"
nx = int(self._cif_header_dictionary["X-Binary-Size-Fastest-Dimension"])
ny = int(self._cif_header_dictionary["X-Binary-Size-Second-Dimension"])
if "Count_cutoff" in self._cif_header_dictionary:
overload = int(self._cif_header_dictionary["Count_cutoff"].split()[0])
else:
# missing from data transformed with GPhL converter - dials#376
overload = 100000000
underload = -1
try:
identifier = self._cif_header_dictionary["Detector"]
except KeyError:
identifier = "Unknown Eiger"
from cctbx.eltbx import attenuation_coefficient
table = attenuation_coefficient.get_table(material)
mu = table.mu_at_angstrom(wavelength) / 10.0
t0 = thickness
detector = self._detector_factory.simple(
"PAD",
distance * 1000.0,
(beam_x * pixel_x * 1000.0, beam_y * pixel_y * 1000.0),
"+x",
"-y",
(1000 * pixel_x, 1000 * pixel_y),
(nx, ny),
(underload, overload),
[],
px_mm=ParallaxCorrectedPxMmStrategy(mu, t0),
mu=mu,
)
for f0, f1, s0, s1 in determine_eiger_mask(detector):
detector[0].add_mask(f0 - 1, s0 - 1, f1, s1)
for panel in detector:
panel.set_thickness(thickness)
panel.set_material(material)
panel.set_identifier(identifier)
panel.set_mu(mu)
return detector
