Coverage for python/lsst/ip/isr/isrMockLSST.py: 16%
465 statements
« prev ^ index » next coverage.py v7.11.0, created at 2025-11-04 02:08 -0800
1# This file is part of ip_isr.
2#
3# Developed for the LSST Data Management System.
4# This product includes software developed by the LSST Project
5# (https://www.lsst.org).
6# See the COPYRIGHT file at the top-level directory of this distribution
7# for details of code ownership.
8#
9# This program is free software: you can redistribute it and/or modify
10# it under the terms of the GNU General Public License as published by
11# the Free Software Foundation, either version 3 of the License, or
12# (at your option) any later version.
13#
14# This program is distributed in the hope that it will be useful,
15# but WITHOUT ANY WARRANTY; without even the implied warranty of
16# MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
17# GNU General Public License for more details.
18#
19# You should have received a copy of the GNU General Public License
20# along with this program. If not, see <https://www.gnu.org/licenses/>.
22__all__ = ["IsrMockLSSTConfig", "IsrMockLSST", "RawMockLSST",
23 "CalibratedRawMockLSST", "ReferenceMockLSST",
24 "BiasMockLSST", "DarkMockLSST", "FlatMockLSST", "FringeMockLSST",
25 "BfKernelMockLSST", "DefectMockLSST", "CrosstalkCoeffMockLSST",
26 "TransmissionMockLSST"]
28import numpy as np
29import galsim
31import lsst.geom as geom
32import lsst.pex.config as pexConfig
33import lsst.afw.detection as afwDetection
34import lsst.afw.math as afwMath
35from lsst.afw.cameraGeom import ReadoutCorner
36from lsst.daf.base import PropertyList
37from .crosstalk import CrosstalkCalib
38from .isrMock import IsrMockConfig, IsrMock
39from .defects import Defects
40from .assembleCcdTask import AssembleCcdTask
41from .linearize import Linearizer
42from .brighterFatterKernel import BrighterFatterKernel
43from .deferredCharge import (
44 SegmentSimulator,
45 FloatingOutputAmplifier,
46 DeferredChargeCalib
47)
50class IsrMockLSSTConfig(IsrMockConfig):
51 """Configuration parameters for isrMockLSST.
52 """
53 # Detector parameters and "Exposure" parameters,
54 # mostly inherited from IsrMockConfig.
55 isLsstLike = pexConfig.Field(
56 dtype=bool,
57 default=True,
58 doc="If True, products have one raw image per amplifier, otherwise, one raw image per detector.",
59 )
60 calibMode = pexConfig.Field(
61 dtype=bool,
62 default=False,
63 doc="Set to true to produce mock calibration products, e.g. combined bias, dark, flat, etc.",
64 )
65 doAdd2DBias = pexConfig.Field(
66 dtype=bool,
67 default=True,
68 doc="Add 2D bias residual frame to data.",
69 )
70 doAddBrightDefects = pexConfig.Field(
71 dtype=bool,
72 default=True,
73 doc="Add bright defects (bad column) to data.",
74 )
75 brightDefectLevel = pexConfig.Field(
76 dtype=float,
77 default=30000.0,
78 doc="Bright defect level (electron).",
79 )
80 doAddBadParallelOverscanColumn = pexConfig.Field(
81 dtype=bool,
82 default=True,
83 doc="Add a bad column to the parallel overscan.",
84 )
85 badParallelOverscanColumnLevel = pexConfig.Field(
86 dtype=float,
87 default=300000.,
88 doc="Bright parallel overscan column level (electron). Should be above saturation.",
89 )
90 doAddBadParallelOverscanColumnNeighbors = pexConfig.Field(
91 dtype=bool,
92 default=True,
93 doc="Add low-level bad columns next to parallel overscan bad column.",
94 )
95 badParallelOverscanColumnNeighborsLevel = pexConfig.Field(
96 dtype=float,
97 default=50.0,
98 doc="Bright parallel overscan column neighbors level (electron).",
99 )
100 doAddBrighterFatter = pexConfig.Field(
101 dtype=bool,
102 default=False,
103 doc="Add brighter fatter and/or diffusion effects to image.",
104 )
105 doAddDeferredCharge = pexConfig.Field(
106 dtype=bool,
107 default=False,
108 doc="Add serial CTI at the amp level?",
109 )
110 bfStrength = pexConfig.Field(
111 dtype=float,
112 default=2.0,
113 doc="The brighter fatter effect scaling parameter (cannot be zero)."
114 "Nominally = 1, but = 2 is more realistic."
115 )
116 nRecalc = pexConfig.Field(
117 dtype=int,
118 default=10000,
119 doc="Number of electrons to accumulate before recalculating pixel shapes.",
120 )
121 doAddClockInjectedOffset = pexConfig.Field(
122 dtype=bool,
123 default=True,
124 doc="Add clock-injected offset to data (on-chip bias level).",
125 )
126 clockInjectedOffsetLevel = pexConfig.Field(
127 dtype=float,
128 default=8500.0,
129 doc="Clock-injected offset (on-chip bias level), in electron.",
130 )
131 noise2DBias = pexConfig.Field(
132 dtype=float,
133 default=2.0,
134 doc="Noise (in electron) to generate a 2D bias residual frame.",
135 )
136 doAddDarkNoiseOnly = pexConfig.Field(
137 dtype=bool,
138 default=False,
139 doc="Add only dark current noise, for testing consistency.",
140 )
141 doAddParallelOverscanRamp = pexConfig.Field(
142 dtype=bool,
143 default=True,
144 doc="Add overscan ramp to parallel overscan and data regions.",
145 )
146 doAddSerialOverscanRamp = pexConfig.Field(
147 dtype=bool,
148 default=True,
149 doc="Add overscan ramp to serial overscan and data regions.",
150 )
151 doAddHighSignalNonlinearity = pexConfig.Field(
152 dtype=bool,
153 default=True,
154 doc="Add high signal non-linearity to overscan and data regions?",
155 )
156 doAddLowSignalNonlinearity = pexConfig.Field(
157 dtype=bool,
158 default=False,
159 doc="Add low signal non-linearity to overscan and data regions? (Not supported yet.)",
160 )
161 highSignalNonlinearityThreshold = pexConfig.Field(
162 dtype=float,
163 default=40_000.,
164 doc="Threshold (in adu) for the non-linearity to be considered ``high signal``.",
165 )
166 doApplyGain = pexConfig.Field(
167 dtype=bool,
168 default=True,
169 doc="Add gain to data.",
170 )
171 doRoundAdu = pexConfig.Field(
172 dtype=bool,
173 default=True,
174 doc="Round adu values to nearest integer.",
175 )
176 gainDict = pexConfig.DictField(
177 keytype=str,
178 itemtype=float,
179 doc="Dictionary of amp name to gain; any amps not listed will use "
180 "config.gain as the value. Units are electron/adu.",
181 default={
182 "C:0,0": 1.65,
183 "C:0,1": 1.60,
184 "C:0,2": 1.55,
185 "C:0,3": 1.70,
186 "C:1,0": 1.75,
187 "C:1,1": 1.80,
188 "C:1,2": 1.85,
189 "C:1,3": 1.70,
190 },
191 )
192 assembleCcd = pexConfig.ConfigurableField(
193 target=AssembleCcdTask,
194 doc="CCD assembly task; used for defect box conversions.",
195 )
197 def validate(self):
198 super().validate()
200 if self.doAddLowSignalNonlinearity:
201 raise NotImplementedError("Low signal non-linearity is not implemented.")
203 if self.doAddDeferredCharge and self.isTrimmed:
204 raise NotImplementedError("Must be untrimmed for mock serial CTI to"
205 "realistically add charge into overscan regions.")
207 def setDefaults(self):
208 super().setDefaults()
210 self.gain = 1.7 # Default value.
211 self.skyLevel = 1700.0 # electron
212 self.sourceFlux = [50_000.0] # electron
213 self.sourceX = [35.0] # pixel
214 self.sourceY = [37.0] # pixel
215 self.overscanScale = 170.0 # electron
216 self.biasLevel = 20_000.0 # adu
217 self.doAddCrosstalk = True
220class IsrMockLSST(IsrMock):
221 """Class to generate consistent mock images for ISR testing.
222 """
223 ConfigClass = IsrMockLSSTConfig
224 _DefaultName = "isrMockLSST"
226 def __init__(self, **kwargs):
227 super().__init__(**kwargs)
229 # Get kernel derived from imSim generated flats with BFE. The kernel
230 # was used for Ops Rehearsal 3 for LSSTCam-type sensors
231 # See https://rubinobs.atlassian.net/browse/DM-43059 for more details.
232 self.bfKernel = np.array([[4.83499829e-01, 8.10171823e-01, 5.31096720e-01,
233 3.54369868e-02, -8.44782871e-01, -1.64614462e+00,
234 -3.83933101e+00, -5.60243416e+00, -6.51691578e+00,
235 -5.60243416e+00, -3.83933101e+00, -1.64614462e+00,
236 -8.44782871e-01, 3.54369868e-02, 5.31096720e-01,
237 8.10171823e-01, 4.83499829e-01],
238 [1.12382749e+00, 2.22609074e+00, 1.27877807e+00,
239 4.55434098e-01, -1.76842385e+00, -1.90046460e+00,
240 -8.10874526e+00, -1.20534899e+01, -1.48627948e+01,
241 -1.20534899e+01, -8.10874526e+00, -1.90046460e+00,
242 -1.76842385e+00, 4.55434098e-01, 1.27877807e+00,
243 2.22609074e+00, 1.12382749e+00],
244 [1.78571940e+00, 4.38918110e+00, 3.95098587e+00,
245 3.70961649e-01, -3.48151981e+00, -9.61567736e+00,
246 -1.78621172e+01, -2.32278872e+01, -2.31833727e+01,
247 -2.32278872e+01, -1.78621172e+01, -9.61567736e+00,
248 -3.48151981e+00, 3.70961649e-01, 3.95098587e+00,
249 4.38918110e+00, 1.78571940e+00],
250 [1.62986900e+00, 3.67851228e+00, 5.68645252e+00,
251 2.15342566e-01, -8.89937202e+00, -1.44739813e+01,
252 -2.98952660e+01, -4.37420817e+01, -4.83160958e+01,
253 -4.37420817e+01, -2.98952660e+01, -1.44739813e+01,
254 -8.89937202e+00, 2.15342566e-01, 5.68645252e+00,
255 3.67851228e+00, 1.62986900e+00],
256 [1.05524430e+00, 1.71917897e+00, 1.73105590e+00,
257 -2.10088420e+00, -1.15118208e+01, -2.55007598e+01,
258 -4.73056159e+01, -6.97257685e+01, -8.09264433e+01,
259 -6.97257685e+01, -4.73056159e+01, -2.55007598e+01,
260 -1.15118208e+01, -2.10088420e+00, 1.73105590e+00,
261 1.71917897e+00, 1.05524430e+00],
262 [8.71929228e-01, 5.41025574e-01, 9.47560771e-01,
263 -5.75314708e-01, -7.46104027e+00, -4.42314481e+01,
264 -9.54126971e+01, -1.61603201e+02, -2.07520692e+02,
265 -1.61603201e+02, -9.54126971e+01, -4.42314481e+01,
266 -7.46104027e+00, -5.75314708e-01, 9.47560771e-01,
267 5.41025574e-01, 8.71929228e-01],
268 [1.89144704e+00, 3.57543979e+00, -6.91419168e-02,
269 -3.37950835e+00, -1.46695089e+01, -7.22850746e+01,
270 -1.65563055e+02, -3.10820425e+02, -4.70026655e+02,
271 -3.10820425e+02, -1.65563055e+02, -7.22850746e+01,
272 -1.46695089e+01, -3.37950835e+00, -6.91419168e-02,
273 3.57543979e+00, 1.89144704e+00],
274 [3.11841913e+00, 7.84024994e+00, 1.88495248e+00,
275 -7.69011009e+00, -2.71782400e+01, -1.04343326e+02,
276 -2.47561370e+02, -5.32959841e+02, -1.16529012e+03,
277 -5.32959841e+02, -2.47561370e+02, -1.04343326e+02,
278 -2.71782400e+01, -7.69011009e+00, 1.88495248e+00,
279 7.84024994e+00, 3.11841913e+00],
280 [2.74197956e+00, 4.73107997e+00, -9.48352966e-01,
281 -9.44822832e+00, -3.06477671e+01, -1.26788739e+02,
282 -3.22828411e+02, -8.47943472e+02, -3.87702420e+03,
283 -8.47943472e+02, -3.22828411e+02, -1.26788739e+02,
284 -3.06477671e+01, -9.44822832e+00, -9.48352966e-01,
285 4.73107997e+00, 2.74197956e+00],
286 [3.11841913e+00, 7.84024994e+00, 1.88495248e+00,
287 -7.69011009e+00, -2.71782400e+01, -1.04343326e+02,
288 -2.47561370e+02, -5.32959841e+02, -1.16529012e+03,
289 -5.32959841e+02, -2.47561370e+02, -1.04343326e+02,
290 -2.71782400e+01, -7.69011009e+00, 1.88495248e+00,
291 7.84024994e+00, 3.11841913e+00],
292 [1.89144704e+00, 3.57543979e+00, -6.91419168e-02,
293 -3.37950835e+00, -1.46695089e+01, -7.22850746e+01,
294 -1.65563055e+02, -3.10820425e+02, -4.70026655e+02,
295 -3.10820425e+02, -1.65563055e+02, -7.22850746e+01,
296 -1.46695089e+01, -3.37950835e+00, -6.91419168e-02,
297 3.57543979e+00, 1.89144704e+00],
298 [8.71929228e-01, 5.41025574e-01, 9.47560771e-01,
299 -5.75314708e-01, -7.46104027e+00, -4.42314481e+01,
300 -9.54126971e+01, -1.61603201e+02, -2.07520692e+02,
301 -1.61603201e+02, -9.54126971e+01, -4.42314481e+01,
302 -7.46104027e+00, -5.75314708e-01, 9.47560771e-01,
303 5.41025574e-01, 8.71929228e-01],
304 [1.05524430e+00, 1.71917897e+00, 1.73105590e+00,
305 -2.10088420e+00, -1.15118208e+01, -2.55007598e+01,
306 -4.73056159e+01, -6.97257685e+01, -8.09264433e+01,
307 -6.97257685e+01, -4.73056159e+01, -2.55007598e+01,
308 -1.15118208e+01, -2.10088420e+00, 1.73105590e+00,
309 1.71917897e+00, 1.05524430e+00],
310 [1.62986900e+00, 3.67851228e+00, 5.68645252e+00,
311 2.15342566e-01, -8.89937202e+00, -1.44739813e+01,
312 -2.98952660e+01, -4.37420817e+01, -4.83160958e+01,
313 -4.37420817e+01, -2.98952660e+01, -1.44739813e+01,
314 -8.89937202e+00, 2.15342566e-01, 5.68645252e+00,
315 3.67851228e+00, 1.62986900e+00],
316 [1.78571940e+00, 4.38918110e+00, 3.95098587e+00,
317 3.70961649e-01, -3.48151981e+00, -9.61567736e+00,
318 -1.78621172e+01, -2.32278872e+01, -2.31833727e+01,
319 -2.32278872e+01, -1.78621172e+01, -9.61567736e+00,
320 -3.48151981e+00, 3.70961649e-01, 3.95098587e+00,
321 4.38918110e+00, 1.78571940e+00],
322 [1.12382749e+00, 2.22609074e+00, 1.27877807e+00,
323 4.55434098e-01, -1.76842385e+00, -1.90046460e+00,
324 -8.10874526e+00, -1.20534899e+01, -1.48627948e+01,
325 -1.20534899e+01, -8.10874526e+00, -1.90046460e+00,
326 -1.76842385e+00, 4.55434098e-01, 1.27877807e+00,
327 2.22609074e+00, 1.12382749e+00],
328 [4.83499829e-01, 8.10171823e-01, 5.31096720e-01,
329 3.54369868e-02, -8.44782871e-01, -1.64614462+00,
330 -3.83933101e+00, -5.60243416e+00, -6.51691578e+00,
331 -5.60243416e+00, -3.83933101e+00, -1.64614462e+00,
332 -8.44782871e-01, 3.54369868e-02, 5.31096720e-01,
333 8.10171823e-01, 4.83499829e-01]]) * 1e-10
335 # Spline trap coefficients and the ctiCalibDict are all taken from a
336 # cti calibration measured from LSSTCam sensor R03_S12 during Run 5
337 # EO testing. These are the coefficients for the spline trap model
338 # used in the deferred charge calibration. The collection can be
339 # found in /repo/ir2: u/abrought/13144/cti (processed 3/4/2024).
340 self.splineTrapCoeffs = np.array([0.0, 28.1, 47.4, 56.4, 66.6, 78.6, 92.4, 109.4,
341 129.0, 151.9, 179.4, 211.9, 250.5, 296.2, 350.0,
342 413.5, 488.0, 576.0, 680.4, 753.0, 888.2, 1040.5,
343 1254.1, 1478.9, 1747.0, 2055.7, 2416.9, 2855.2,
344 3361.9, 3969.4, 4665.9, 5405.3, 6380.0, 7516.7,
345 8875.9, 10488.6, 12681.9, 14974.2, 17257.6, 20366.5,
346 24026.7, 28372.1, 33451.7, 39550.4, 46624.8, 55042.9,
347 64862.7, 76503.1, 90265.6, 106384.2, 0.0, 0.0, 0.0,
348 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.1,
349 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.1, 0.0,
350 0.1, 0.2, 0.6, 0.1, 0.0, 0.1, 0.0, 0.6, 0.3, 0.5, 0.8,
351 0.8, 1.5, 2.0, 1.8, 2.4, 2.6, 3.7, 5.0, 6.4, 8.4, 10.9,
352 14.5, 21.1, 28.9])
354 self.ctiCalibDict = {'driftScale': {'C:0,0': 0.000127,
355 'C:0,1': 0.000137,
356 'C:0,2': 0.000138,
357 'C:0,3': 0.000147,
358 'C:1,0': 0.000147,
359 'C:1,1': 0.000122,
360 'C:1,2': 0.000123,
361 'C:1,3': 0.000116},
362 'decayTime': {'C:0,0': 2.30,
363 'C:0,1': 2.21,
364 'C:0,2': 2.28,
365 'C:0,3': 2.34,
366 'C:1,0': 2.30,
367 'C:1,1': 2.40,
368 'C:1,2': 2.51,
369 'C:1,3': 2.21},
370 'globalCti': {'C:0,0': 5.25e-07,
371 'C:0,1': 5.38e-07,
372 'C:0,2': 5.80e-07,
373 'C:0,3': 5.91e-07,
374 'C:1,0': 6.24e-07,
375 'C:1,1': 5.72e-07,
376 'C:1,2': 5.60e-07,
377 'C:1,3': 4.40e-07},
378 'signals': {'C:0,0': np.linspace(1.0e2, 1.0e5, 10),
379 'C:0,1': np.linspace(1.0e2, 1.0e5, 10),
380 'C:0,2': np.linspace(1.0e2, 1.0e5, 10),
381 'C:0,3': np.linspace(1.0e2, 1.0e5, 10),
382 'C:1,0': np.linspace(1.0e2, 1.0e5, 10),
383 'C:1,1': np.linspace(1.0e2, 1.0e5, 10),
384 'C:1,2': np.linspace(1.0e2, 1.0e5, 10),
385 'C:1,3': np.linspace(1.0e2, 1.0e5, 10)},
386 'serialEper': {'C:0,0': np.full(10, 5.25e-07),
387 'C:0,1': np.full(10, 5.38e-07),
388 'C:0,2': np.full(10, 5.80e-07),
389 'C:0,3': np.full(10, 5.91e-07),
390 'C:1,0': np.full(10, 6.24e-07),
391 'C:1,1': np.full(10, 5.72e-07),
392 'C:1,2': np.full(10, 5.60e-07),
393 'C:1,3': np.full(10, 4.40e-07)},
394 'parallelEper': {'C:0,0': np.full(10, 5.25e-07),
395 'C:0,1': np.full(10, 5.38e-07),
396 'C:0,2': np.full(10, 5.80e-07),
397 'C:0,3': np.full(10, 5.91e-07),
398 'C:1,0': np.full(10, 6.24e-07),
399 'C:1,1': np.full(10, 5.72e-07),
400 'C:1,2': np.full(10, 5.60e-07),
401 'C:1,3': np.full(10, 4.40e-07)},
402 'serialCtiTurnoff': {'C:0,0': 1.0e5,
403 'C:0,1': 1.0e5,
404 'C:0,2': 1.0e5,
405 'C:0,3': 1.0e5,
406 'C:1,0': 1.0e5,
407 'C:1,1': 1.0e5,
408 'C:1,2': 1.0e5,
409 'C:1,3': 1.0e5},
410 'parallelCtiTurnoff': {'C:0,0': 1.0e5,
411 'C:0,1': 1.0e5,
412 'C:0,2': 1.0e5,
413 'C:0,3': 1.0e5,
414 'C:1,0': 1.0e5,
415 'C:1,1': 1.0e5,
416 'C:1,2': 1.0e5,
417 'C:1,3': 1.0e5},
418 'serialCtiTurnoffSamplingErr': {'C:0,0': 1.0e3,
419 'C:0,1': 1.0e3,
420 'C:0,2': 1.0e3,
421 'C:0,3': 1.0e3,
422 'C:1,0': 1.0e3,
423 'C:1,1': 1.0e3,
424 'C:1,2': 1.0e3,
425 'C:1,3': 1.0e3},
426 'parallelCtiTurnoffSamplingErr': {'C:0,0': 1.0e3,
427 'C:0,1': 1.0e3,
428 'C:0,2': 1.0e3,
429 'C:0,3': 1.0e3,
430 'C:1,0': 1.0e3,
431 'C:1,1': 1.0e3,
432 'C:1,2': 1.0e3,
433 'C:1,3': 1.0e3},
434 'serialTraps': {'C:0,0': {'size': 20000.0,
435 'emissionTime': 0.4,
436 'pixel': 1,
437 'trap_type': 'spline',
438 'coeffs': self.splineTrapCoeffs},
439 'C:0,1': {'size': 20000.0,
440 'emissionTime': 0.4,
441 'pixel': 1,
442 'trap_type': 'spline',
443 'coeffs': self.splineTrapCoeffs},
444 'C:0,2': {'size': 20000.0,
445 'emissionTime': 0.4,
446 'pixel': 1,
447 'trap_type': 'spline',
448 'coeffs': self.splineTrapCoeffs},
449 'C:0,3': {'size': 20000.0,
450 'emissionTime': 0.4,
451 'pixel': 1,
452 'trap_type': 'spline',
453 'coeffs': self.splineTrapCoeffs},
454 'C:1,0': {'size': 20000.0,
455 'emissionTime': 0.4,
456 'pixel': 1,
457 'trap_type': 'spline',
458 'coeffs': self.splineTrapCoeffs},
459 'C:1,1': {'size': 20000.0,
460 'emissionTime': 0.4,
461 'pixel': 1,
462 'trap_type': 'spline',
463 'coeffs': self.splineTrapCoeffs},
464 'C:1,2': {'size': 20000.0,
465 'emissionTime': 0.4,
466 'pixel': 1,
467 'trap_type': 'spline',
468 'coeffs': self.splineTrapCoeffs},
469 'C:1,3': {'size': 20000.0,
470 'emissionTime': 0.4,
471 'pixel': 1,
472 'trap_type': 'spline',
473 'coeffs': self.splineTrapCoeffs}}}
475 self.deferredChargeCalib = self.makeDeferredChargeCalib()
477 # Cross-talk coeffs are defined in the parent class.
479 self.makeSubtask("assembleCcd")
481 def run(self):
482 """Generate a mock ISR product following LSSTCam ISR, and return it.
484 Returns
485 -------
486 image : `lsst.afw.image.Exposure`
487 Simulated ISR image with signals added.
488 dataProduct :
489 Simulated ISR data products.
490 None :
491 Returned if no valid configuration was found.
493 Raises
494 ------
495 RuntimeError
496 Raised if both doGenerateImage and doGenerateData are specified.
497 """
498 if self.config.doGenerateImage and self.config.doGenerateData:
499 raise RuntimeError("Only one of doGenerateImage and doGenerateData may be specified.")
500 elif self.config.doGenerateImage:
501 return self.makeImage()
502 elif self.config.doGenerateData:
503 return self.makeData()
504 else:
505 return None
507 def makeImage(self):
508 """Generate a simulated ISR LSST image.
510 Returns
511 -------
512 exposure : `lsst.afw.image.Exposure` or `dict`
513 Simulated ISR image data.
515 Notes
516 -----
517 This method constructs a "raw" data image.
518 """
519 exposure = self.getExposure()
521 # Set up random number generators for consistency of components,
522 # no matter the group that are configured.
523 rngSky = np.random.RandomState(seed=self.config.rngSeed + 1)
524 rngDark = np.random.RandomState(seed=self.config.rngSeed + 2)
525 rng2DBias = np.random.RandomState(seed=self.config.rngSeed + 3)
526 rngOverscan = np.random.RandomState(seed=self.config.rngSeed + 4)
527 rngReadNoise = np.random.RandomState(seed=self.config.rngSeed + 5)
528 rngBrighterFatter = galsim.BaseDeviate(self.config.rngSeed + 6)
530 # Create the linearizer if we will need it.
531 if self.config.doAddHighSignalNonlinearity:
532 linearizer = LinearizerMockLSST().run()
534 # We introduce effects as they happen from a source to the signal,
535 # so the effects go from electron to adu.
536 # The ISR steps will then correct these effects in the reverse order.
537 for idx, amp in enumerate(exposure.getDetector()):
538 # Get image bbox and data
539 bbox = None
540 if self.config.isTrimmed:
541 bbox = amp.getBBox()
542 bboxFull = bbox
543 else:
544 bbox = amp.getRawDataBBox()
545 bboxFull = amp.getRawBBox()
547 # This is the image data (excluding pre/overscans).
548 ampImageData = exposure.image[bbox]
549 # This is the full data (including pre/overscans if untrimmed).
550 ampFullData = exposure.image[bboxFull]
552 # Astrophysical signals are all in electron (e-).
553 # These are only applied to the imaging portion of the
554 # amplifier (ampImageData)
556 if self.config.doAddSky:
557 # The sky effects are in electron.
558 self.amplifierAddNoise(
559 ampImageData,
560 self.config.skyLevel,
561 np.sqrt(self.config.skyLevel),
562 rng=rngSky,
563 )
565 if self.config.doAddSource:
566 for sourceAmp, sourceFlux, sourceX, sourceY in zip(self.config.sourceAmp,
567 self.config.sourceFlux,
568 self.config.sourceX,
569 self.config.sourceY):
570 if idx == sourceAmp:
571 # The source flux is in electron.
572 self.amplifierAddSource(ampImageData, sourceFlux, sourceX, sourceY)
574 if self.config.doAddFringe:
575 # Fringes are added in electron.
576 self.amplifierAddFringe(amp,
577 ampImageData,
578 np.array(self.config.fringeScale),
579 x0=np.array(self.config.fringeX0),
580 y0=np.array(self.config.fringeY0))
582 if self.config.doAddFlat:
583 if self.config.calibMode:
584 # In case we are making a combined flat,
585 # add a non-zero signal so the mock flat can be multiplied
586 self.amplifierAddNoise(ampImageData, 1.0, 0.0)
587 # Multiply each amplifier by a Gaussian centered on u0 and v0
588 u0 = exposure.getDetector().getBBox().getDimensions().getX()/2.
589 v0 = exposure.getDetector().getBBox().getDimensions().getY()/2.
590 self.amplifierMultiplyFlat(amp, ampImageData, self.config.flatDrop, u0=u0, v0=v0)
592 # On-chip electronic effects.
594 # 1. Add bright defect(s).
595 if self.config.doAddBrightDefects:
596 defectList = self.makeDefectList(isTrimmed=self.config.isTrimmed)
598 for defect in defectList:
599 exposure.image[defect.getBBox()] = self.config.brightDefectLevel
601 for idx, amp in enumerate(exposure.getDetector()):
602 # Get image bbox and data
603 bbox = None
604 if self.config.isTrimmed:
605 bbox = amp.getBBox()
606 bboxFull = bbox
607 else:
608 bbox = amp.getRawDataBBox()
609 bboxFull = amp.getRawBBox()
611 # This is the image data (excluding pre/overscans).
612 ampImageData = exposure.image[bbox]
613 # This is the full data (including pre/overscans if untrimmed).
614 ampFullData = exposure.image[bboxFull]
616 # 2. Add dark current (electron) to imaging portion of the amp.
617 if self.config.doAddDark or self.config.doAddDarkNoiseOnly:
618 if self.config.doAddDarkNoiseOnly:
619 darkLevel = 0.0
620 else:
621 darkLevel = self.config.darkRate * self.config.darkTime
622 if self.config.calibMode:
623 darkNoise = 0.0
624 else:
625 darkNoise = np.sqrt(self.config.darkRate * self.config.darkTime)
627 self.amplifierAddNoise(ampImageData, darkLevel, darkNoise, rng=rngDark)
629 # 3. Add BF effect (electron) to imaging portion of the amp.
630 if self.config.doAddBrighterFatter is True:
631 self.amplifierAddBrighterFatter(
632 ampImageData,
633 rngBrighterFatter,
634 self.config.bfStrength,
635 self.config.nRecalc,
636 )
638 # 4. Add serial CTI (electron) to amplifier (imaging + overscan).
639 if self.config.doAddDeferredCharge:
640 # Get the free charge area for the amplifier.
641 self.amplifierAddDeferredCharge(exposure, amp)
643 # 5. Add 2D bias residual (electron) to imaging portion of the amp.
644 if self.config.doAdd2DBias:
645 # For now we use an unstructured noise field to add some
646 # consistent 2D bias residual that can be subtracted. In
647 # the future this can be made into a warm corner (for example).
648 self.amplifierAddNoise(
649 ampImageData,
650 0.0,
651 self.config.noise2DBias,
652 rng=rng2DBias,
653 )
655 # 6. Add clock-injected offset (electron) to amplifer
656 # (imaging + overscan).
657 # This is just an offset that will be crosstalked and modified by
658 # the gain, and does not have a noise associated with it.
659 if self.config.doAddClockInjectedOffset:
660 self.amplifierAddNoise(
661 ampFullData,
662 self.config.clockInjectedOffsetLevel,
663 0.0,
664 )
666 # 7./8. Add serial and parallel overscan slopes (electron)
667 # (imaging + overscan)
668 if (self.config.doAddParallelOverscanRamp or self.config.doAddSerialOverscanRamp) and \
669 not self.config.isTrimmed:
671 if self.config.doAddParallelOverscanRamp:
672 # Apply gradient along the X axis.
673 self.amplifierAddXGradient(ampFullData, -1.0 * self.config.overscanScale,
674 1.0 * self.config.overscanScale)
676 if self.config.doAddSerialOverscanRamp:
677 # Apply the gradient along the Y axis.
678 self.amplifierAddYGradient(ampFullData, -1.0 * self.config.overscanScale,
679 1.0 * self.config.overscanScale)
681 # 9. Add non-linearity (electron) to amplifier
682 # (imaging + overscan).
683 if self.config.doAddHighSignalNonlinearity:
684 # The linearizer coefficients come from makeLinearizer().
685 if linearizer.linearityType[amp.getName()] != "Spline":
686 raise RuntimeError("IsrMockLSST only supports spline non-linearity.")
688 coeffs = linearizer.linearityCoeffs[amp.getName()]
689 centers, values = np.split(coeffs, 2)
691 # This is an application of high signal non-linearity, so we
692 # set the lower values to 0.0 (this cut is arbitrary).
693 values[centers < self.config.highSignalNonlinearityThreshold] = 0.0
695 # The linearizer is units of adu, so convert to electron
696 centers *= self.config.gainDict[amp.getName()]
697 values *= self.config.gainDict[amp.getName()]
699 # Note that the linearity spline is in "overscan subtracted"
700 # units so needs to be applied without the clock-injected
701 # offset.
702 self.amplifierAddNonlinearity(
703 ampFullData,
704 centers,
705 values,
706 self.config.clockInjectedOffsetLevel if self.config.doAddClockInjectedOffset else 0.0,
707 )
709 # 10. Add read noise (electron) to the amplifier
710 # (imaging + overscan).
711 # Unsure if this should be before or after crosstalk.
712 # Probably some of both; hopefully doesn't matter.
713 if not self.config.calibMode:
714 # Add read noise to the imaging region.
715 self.amplifierAddNoise(
716 ampImageData,
717 0.0,
718 self.config.readNoise,
719 rng=rngReadNoise,
720 )
722 # If not trimmed, add to the overscan regions.
723 if not self.config.isTrimmed:
724 parallelOverscanBBox = amp.getRawParallelOverscanBBox()
725 parallelOverscanData = exposure.image[parallelOverscanBBox]
727 serialOverscanBBox = self.getFullSerialOverscanBBox(amp)
728 serialOverscanData = exposure.image[serialOverscanBBox]
730 # Add read noise of mean 0
731 # to the parallel and serial overscan regions.
732 self.amplifierAddNoise(
733 parallelOverscanData,
734 0.0,
735 self.config.readNoise,
736 rng=rngOverscan,
737 )
738 self.amplifierAddNoise(
739 serialOverscanData,
740 0.0,
741 self.config.readNoise,
742 rng=rngOverscan,
743 )
745 # 7b. Add bad column to the parallel overscan region.
746 if self.config.doAddBadParallelOverscanColumn and not self.config.isTrimmed:
747 # We want to place this right above the defect, to simulate
748 # bleeding into the parallel overscan region.
749 amp = exposure.getDetector()[2]
750 parBBox = amp.getRawParallelOverscanBBox()
751 bboxBad = geom.Box2I(
752 corner=geom.Point2I(50, parBBox.getMinY()),
753 dimensions=geom.Extent2I(1, parBBox.getHeight()),
754 )
755 exposure[bboxBad].image.array[:, :] += self.config.badParallelOverscanColumnLevel
757 if self.config.doAddBadParallelOverscanColumnNeighbors:
758 for neighbor in [49, 51]:
759 bboxBad = geom.Box2I(
760 corner=geom.Point2I(neighbor, parBBox.getMinY()),
761 dimensions=geom.Extent2I(1, parBBox.getHeight()),
762 )
763 exposure[bboxBad].image.array[:, :] += self.config.badParallelOverscanColumnNeighborsLevel
765 # 11. Add crosstalk (electron) to all the amplifiers
766 # (imaging + overscan).
767 if self.config.doAddCrosstalk:
768 ctCalib = CrosstalkCalib()
769 # We use the regular subtractCrosstalk code but with a negative
770 # sign on the crosstalk coefficients so it adds instead of
771 # subtracts. We only apply the signal plane (ignoreVariance,
772 # subtrahendMasking) with a very large pixel to mask to ensure
773 # no crosstalk mask bits are set.
774 ctCalib.subtractCrosstalk(
775 exposure,
776 crosstalkCoeffs=-1*self.crosstalkCoeffs,
777 doSubtrahendMasking=True,
778 minPixelToMask=np.inf,
779 ignoreVariance=True,
780 fullAmplifier=True,
781 )
783 for amp in exposure.getDetector():
784 # Get image bbox and data (again).
785 bbox = None
786 if self.config.isTrimmed:
787 bbox = amp.getBBox()
788 bboxFull = bbox
789 else:
790 bbox = amp.getRawDataBBox()
791 bboxFull = amp.getRawBBox()
793 # This is the image data (excluding pre/overscans).
794 ampImageData = exposure.image[bbox]
795 # This is the full data (including pre/overscans if untrimmed).
796 ampFullData = exposure.image[bboxFull]
798 # 12. Gain un-normalize (from electron to floating point adu)
799 if self.config.doApplyGain:
800 gain = self.config.gainDict.get(amp.getName(), self.config.gain)
801 self.applyGain(ampFullData, gain)
803 # 13. Add overall bias level (adu) to the amplifier
804 # (imaging + overscan)
805 if self.config.doAddBias:
806 self.addBiasLevel(ampFullData, self.config.biasLevel)
808 # 14. Round/Truncate to integers (adu)
809 if self.config.doRoundAdu:
810 self.roundADU(ampFullData)
812 # Add units metadata to calibrations.
813 if self.config.calibMode:
814 if self.config.doApplyGain:
815 exposure.metadata["LSST ISR UNITS"] = "adu"
816 else:
817 exposure.metadata["LSST ISR UNITS"] = "electron"
819 # Add a variance plane appropriate for a calibration frame.
820 # We take the absolute value for biases which have no signal.
821 exposure.variance.array[:, :] = np.abs(np.median(exposure.image.array)/10.)
823 exposure.metadata["BSSVBS"] = 50.0
824 exposure.metadata["HVBIAS"] = "ON"
826 if self.config.doGenerateAmpDict:
827 expDict = dict()
828 for amp in exposure.getDetector():
829 expDict[amp.getName()] = exposure
830 return expDict
831 else:
832 return exposure
834 def addBiasLevel(self, ampData, biasLevel):
835 """Add bias level to an amplifier's image data.
837 Parameters
838 ----------
839 ampData : `lsst.afw.image.ImageF`
840 Amplifier image to operate on.
841 biasLevel : `float`
842 Bias level to be added to the image.
843 """
844 ampArr = ampData.array
845 ampArr[:] = ampArr[:] + biasLevel
847 def makeDefectList(self, isTrimmed=True):
848 """Generate a simple defect list.
850 Parameters
851 ----------
852 isTrimmed : `bool`, optional
853 Return defects in trimmed coordinates?
855 Returns
856 -------
857 defectList : `lsst.meas.algorithms.Defects`
858 Simulated defect list
859 """
860 defectBoxesUntrimmed = [
861 geom.Box2I(
862 geom.Point2I(50, 118),
863 geom.Extent2I(1, 51),
864 ),
865 ]
867 if not isTrimmed:
868 return Defects(defectBoxesUntrimmed)
870 # If trimmed, we need to convert.
871 tempExp = self.getExposure(isTrimmed=False)
872 tempExp.image.array[:, :] = 0.0
873 for bbox in defectBoxesUntrimmed:
874 tempExp.image[bbox] = 1.0
876 assembledExp = self.assembleCcd.assembleCcd(tempExp)
878 # Use thresholding code to find defect footprints/boxes.
879 threshold = afwDetection.createThreshold(1.0, "value", polarity=True)
880 footprintSet = afwDetection.FootprintSet(assembledExp.image, threshold)
882 return Defects.fromFootprintList(footprintSet.getFootprints())
884 def makeBfKernel(self):
885 """Generate a simple simulated brighter-fatter kernel.
886 Returns
887 -------
888 kernel : `lsst.ip.isr.BrighterFatterKernel`
889 Simulated brighter-fatter kernel.
890 """
891 bfkArray = super().makeBfKernel()
892 bfKernelObject = BrighterFatterKernel()
893 bfKernelObject.level = 'AMP'
894 bfKernelObject.gain = self.config.gainDict
896 for amp in self.getExposure().getDetector():
897 # Kernel must be in (y,x) orientation
898 bfKernelObject.ampKernels[amp.getName()] = bfkArray.T
899 bfKernelObject.valid[amp.getName()] = True
901 return bfKernelObject
903 def makeDeferredChargeCalib(self):
904 """Generate a CTI calibration.
906 Returns
907 -------
908 cti : `lsst.ip.isr.deferredCharge.DeferredChargeCalib`
909 Simulated deferred charge calibration.
910 """
912 metadataDict = {'metadata': PropertyList()}
913 metadataDict['metadata'].add(name="OBSTYPE", value="CTI")
914 metadataDict['metadata'].add(name="CALIBCLS",
915 value="lsst.ip.isr.deferredCharge.DeferredChargeCalib")
916 self.ctiCalibDict = {**metadataDict, **self.ctiCalibDict}
917 deferredChargeCalib = DeferredChargeCalib()
918 self.cti = deferredChargeCalib.fromDict(self.ctiCalibDict)
920 return self.cti
922 def amplifierAddBrighterFatter(self, ampImageData, rng, bfStrength, nRecalc):
923 """Add brighter fatter effect and/or diffusion to the image.
924 Parameters
925 ----------
926 ampImageData : `lsst.afw.image.ImageF`
927 Amplifier image to operate on.
928 rng : `galsim.BaseDeviate`
929 Random number generator.
930 bfStrength : `float`
931 Scaling parameter of the brighter fatter effect (nominally = 1)
932 nRecalc: 'int'
933 The number of electrons to accumulate before recalculating the
934 distortion of the pixel shapes.
935 """
937 incidentImage = galsim.Image(ampImageData.array, scale=1)
938 measuredImage = galsim.ImageF(
939 ampImageData.array.shape[1],
940 ampImageData.array.shape[0],
941 scale=1,
942 )
943 photons = galsim.PhotonArray.makeFromImage(incidentImage)
945 sensorModel = galsim.SiliconSensor(
946 strength=bfStrength,
947 rng=rng,
948 diffusion_factor=0.0,
949 nrecalc=nRecalc,
950 )
952 totalFluxAdded = sensorModel.accumulate(photons, measuredImage)
953 ampImageData.array = measuredImage.array
955 return totalFluxAdded
957 def amplifierAddDeferredCharge(self, exposure, amp):
958 """Add serial CTI to the amplifier data.
960 Parameters
961 ----------
962 exposure : `lsst.afw.image.ExposureF`
963 The exposure object containing the amplifier
964 to apply deferred charge to.
965 amp : `lsst.afw.image.Amplifier`
966 The amplifier object (contains geometry info).
967 """
968 # Get the amplifier's geometry parameters.
969 # When adding deferred charge, we have already assured that
970 # isTrimmed is False. Therefore we want to make sure that we
971 # get the RawDataBBox.
972 readoutCorner = amp.getReadoutCorner()
973 prescanWidth = amp.getRawHorizontalPrescanBBox().getWidth()
974 serialOverscanWidth = amp.getRawHorizontalOverscanBBox().getWidth()
975 parallelOverscanWidth = amp.getRawVerticalOverscanBBox().getHeight()
976 bboxFreeCharge = amp.getRawDataBBox()
977 bboxFreeCharge = bboxFreeCharge.expandedTo(amp.getRawHorizontalOverscanBBox())
978 bboxFreeCharge = bboxFreeCharge.expandedTo(amp.getRawHorizontalPrescanBBox())
979 bboxFreeCharge = bboxFreeCharge.expandedTo(amp.getRawVerticalOverscanBBox())
981 ampFreeChargeData = exposure.image[bboxFreeCharge]
982 ampImageData = exposure.image[amp.getRawDataBBox()]
984 # Get the deferred charge parameters for this amplifier.
985 cti = self.deferredChargeCalib.globalCti[amp.getName()]
986 traps = self.deferredChargeCalib.serialTraps[amp.getName()]
987 driftScale = self.deferredChargeCalib.driftScale[amp.getName()]
988 decayTime = self.deferredChargeCalib.decayTime[amp.getName()]
990 # Create a fake amplifier object that contains some deferred charge
991 # paramters.
992 floatingOutputAmplifier = FloatingOutputAmplifier(
993 gain=1.0, # Everything is already in electrons.
994 scale=driftScale,
995 decay_time=decayTime,
996 noise=0.0,
997 offset=0.0,
998 )
1000 def flipImage(arr, readoutCorner):
1001 # Flip an array so that the readout corner is in
1002 # the lower left.
1003 if readoutCorner == ReadoutCorner.LR:
1004 return np.fliplr(arr)
1005 elif readoutCorner == ReadoutCorner.UR:
1006 return np.fliplr(np.flipud(arr))
1007 elif readoutCorner == ReadoutCorner.UL:
1008 return np.flipud(arr)
1009 else:
1010 pass
1012 return arr
1014 # The algorithm expects that the readout corner is in
1015 # the lower left corner. Flip it to be so:
1016 imageData = ampImageData.array
1017 imageData = flipImage(imageData, readoutCorner)
1019 # Simulate the amplifier.
1020 ampSim = SegmentSimulator(
1021 imarr=imageData,
1022 prescan_width=prescanWidth,
1023 output_amplifier=floatingOutputAmplifier,
1024 cti=cti,
1025 traps=traps,
1026 )
1028 # Simulate deferred charge!
1029 # Note that the readout() method uses the image region data and the
1030 # overscan dimensions provided as input parameters. It then creates
1031 # overscan nd adds it to the image data to create the raw image.
1032 result = ampSim.readout(
1033 serial_overscan_width=serialOverscanWidth,
1034 parallel_overscan_width=parallelOverscanWidth,
1035 )
1037 # Flip the image back to the original orientation.
1038 result = flipImage(result, readoutCorner)
1040 # Set the image with the deferred charge added.
1041 ampFreeChargeData.array[:, :] = result
1043 def makeLinearizer(self):
1044 # docstring inherited.
1046 # The linearizer has units of adu.
1047 nNodes = 10
1048 # Set this to just above the mock saturation (adu)
1049 maxADU = 101_000
1050 nonLinSplineNodes = np.linspace(0, maxADU, nNodes)
1051 # These values come from cp_pipe/tests/test_linearity.py and
1052 # are based on a test fit to LSSTCam data, run 7193D, detector 22,
1053 # amp C00.
1054 nonLinSplineValues = np.array(
1055 [0.0, -8.87, 1.46, 1.69, -6.92, -68.23, -78.01, -11.56, 80.26, 185.01]
1056 )
1058 if self.config.doAddHighSignalNonlinearity and not self.config.doAddLowSignalNonlinearity:
1059 nonLinSplineValues[nonLinSplineNodes < self.config.highSignalNonlinearityThreshold] = 0.0
1060 elif self.config.doAddLowSignalNonlinearity:
1061 raise NotImplementedError("Low signal non-linearity is not implemented.")
1063 exp = self.getExposure()
1064 detector = exp.getDetector()
1066 linearizer = Linearizer(detector=detector)
1067 linearizer.updateMetadataFromExposures([exp])
1069 # We need to set override by hand because we are constructing a
1070 # linearizer manually and not from a serialized object.
1071 linearizer.override = True
1072 linearizer.hasLinearity = True
1073 linearizer.validate()
1074 linearizer.updateMetadata(camera=self.getCamera(), detector=detector, filterName='NONE')
1075 linearizer.updateMetadata(setDate=True, setCalibId=True)
1077 for amp in detector:
1078 ampName = amp.getName()
1079 linearizer.linearityType[ampName] = "Spline"
1080 linearizer.linearityCoeffs[ampName] = np.concatenate([nonLinSplineNodes, nonLinSplineValues])
1081 # We need to specify the raw bbox here.
1082 linearizer.linearityBBox[ampName] = amp.getRawBBox()
1084 return linearizer
1086 def amplifierAddNonlinearity(self, ampData, centers, values, offset):
1087 """Add non-linearity to amplifier data.
1089 Parameters
1090 ----------
1091 ampData : `lsst.afw.image.ImageF`
1092 Amplifier image to operate on.
1093 centers : `np.ndarray`
1094 Spline nodes.
1095 values : `np.ndarray`
1096 Spline values.
1097 offset : `float`
1098 Offset zero-point between linearizer (internal vs external).
1099 """
1100 # I'm not sure what to do about negative values...
1102 # Note that we are using the afw AKIMA_SPLINE to offset the
1103 # data but using the equivalent but faster scipy Akima1DInterpolator to
1104 # correct the data.
1105 spl = afwMath.makeInterpolate(
1106 centers,
1107 values,
1108 afwMath.stringToInterpStyle("AKIMA_SPLINE"),
1109 )
1111 delta = np.asarray(spl.interpolate(ampData.array.ravel() - offset))
1113 ampData.array[:, :] += delta.reshape(ampData.array.shape)
1115 def amplifierMultiplyFlat(self, amp, ampData, fracDrop, u0=100.0, v0=100.0):
1116 """Multiply an amplifier's image data by a flat-like pattern.
1118 Parameters
1119 ----------
1120 amp : `lsst.afw.ampInfo.AmpInfoRecord`
1121 Amplifier to operate on. Needed for amp<->exp coordinate
1122 transforms.
1123 ampData : `lsst.afw.image.ImageF`
1124 Amplifier image to operate on.
1125 fracDrop : `float`
1126 Fractional drop from center to edge of detector along x-axis.
1127 u0 : `float`
1128 Peak location in detector coordinates.
1129 v0 : `float`
1130 Peak location in detector coordinates.
1131 """
1132 if fracDrop >= 1.0:
1133 raise RuntimeError("Flat fractional drop cannot be greater than 1.0")
1135 sigma = u0 / np.sqrt(2.0 * fracDrop)
1137 for x in range(0, ampData.getDimensions().getX()):
1138 for y in range(0, ampData.getDimensions().getY()):
1139 (u, v) = self.localCoordToExpCoord(amp, x, y)
1140 f = np.exp(-0.5 * ((u - u0)**2 + (v - v0)**2) / sigma**2)
1141 ampData.array[y][x] = (ampData.array[y][x] * f)
1143 def applyGain(self, ampData, gain):
1144 """Apply gain to the amplifier's data.
1145 This method divides the data by the gain
1146 because the mocks need to convert the data in electron to adu,
1147 so it does the inverse operation to applyGains in isrFunctions.
1149 Parameters
1150 ----------
1151 ampData : `lsst.afw.image.ImageF`
1152 Amplifier image to operate on.
1153 gain : `float`
1154 Gain value in electron/adu.
1155 """
1156 ampArr = ampData.array
1157 ampArr[:] = ampArr[:] / gain
1159 def roundADU(self, ampData):
1160 """Round adu to nearest integer.
1162 Parameters
1163 ----------
1164 ampData : `lsst.afw.image.ImageF`
1165 Amplifier image to operate on.
1166 """
1167 ampArr = ampData.array
1168 ampArr[:] = np.around(ampArr)
1170 def amplifierAddXGradient(self, ampData, start, end):
1171 """Add a x-axis linear gradient to an amplifier's image data.
1173 This method operates in the amplifier coordinate frame.
1175 Parameters
1176 ----------
1177 ampData : `lsst.afw.image.ImageF`
1178 Amplifier image to operate on.
1179 start : `float`
1180 Start value of the gradient (at x=0).
1181 end : `float`
1182 End value of the gradient (at x=xmax).
1183 """
1184 nPixX = ampData.getDimensions().getX()
1185 ampArr = ampData.array
1186 ampArr[:] = ampArr[:] + (np.interp(range(nPixX), (0, nPixX - 1), (start, end)).reshape(1, nPixX)
1187 + np.zeros(ampData.getDimensions()).transpose())
1189 def getFullSerialOverscanBBox(self, amp):
1190 """Get the full serial overscan bounding box from an amplifier.
1192 This includes the serial/parallel overscan region.
1194 Parameters
1195 ----------
1196 amp : `lsst.afw.ampInfo.AmpInfoRecord`
1197 Amplifier to operate on.
1199 Returns
1200 -------
1201 bbox : `lsst.geom.Box2I`
1202 """
1203 # This only works for untrimmed data.
1204 bbox = amp.getRawDataBBox()
1206 parallelOverscanBBox = amp.getRawParallelOverscanBBox()
1207 grownImageBBox = bbox.expandedTo(parallelOverscanBBox)
1209 serialOverscanBBox = amp.getRawSerialOverscanBBox()
1210 # Extend the serial overscan bbox to include corners
1211 serialOverscanBBox = geom.Box2I(
1212 geom.Point2I(serialOverscanBBox.getMinX(),
1213 grownImageBBox.getMinY()),
1214 geom.Extent2I(serialOverscanBBox.getWidth(),
1215 grownImageBBox.getHeight()))
1217 return serialOverscanBBox
1220class RawMockLSST(IsrMockLSST):
1221 """Generate a raw exposure suitable for ISR.
1222 """
1223 def __init__(self, **kwargs):
1224 super().__init__(**kwargs)
1225 self.config.isTrimmed = False
1226 self.config.doGenerateImage = True
1227 self.config.doGenerateAmpDict = False
1229 # Add astro effects
1230 self.config.doAddSky = True
1231 self.config.doAddSource = True
1233 # Add optical effects
1234 self.config.doAddFringe = True
1236 # Add instrument effects
1237 self.config.doAddParallelOverscanRamp = True
1238 self.config.doAddSerialOverscanRamp = True
1239 self.config.doAddCrosstalk = True
1240 self.config.doAddBias = True
1241 self.config.doAddDark = True
1243 self.config.doAddFlat = True
1246class TrimmedRawMockLSST(RawMockLSST):
1247 """Generate a trimmed raw exposure.
1248 """
1249 def __init__(self, **kwargs):
1250 super().__init__(**kwargs)
1251 self.config.isTrimmed = True
1252 self.config.doAddParallelOverscanRamp = False
1253 self.config.doAddSerialOverscanRamp = False
1256class CalibratedRawMockLSST(RawMockLSST):
1257 """Generate a trimmed raw exposure.
1259 This represents a "truth" image that can be compared to a
1260 post-ISR cleaned image.
1261 """
1262 def __init__(self, **kwargs):
1263 super().__init__(**kwargs)
1264 self.config.isTrimmed = True
1265 self.config.doGenerateImage = True
1267 self.config.doAddSky = True
1268 self.config.doAddSource = True
1270 self.config.doAddFringe = True
1272 self.config.doAddParallelOverscanRamp = False
1273 self.config.doAddSerialOverscanRamp = False
1274 self.config.doAddCrosstalk = False
1275 self.config.doAddBias = False
1276 self.config.doAdd2DBias = False
1277 self.config.doAddDark = False
1278 self.config.doApplyGain = False
1279 self.config.doAddFlat = False
1280 self.config.doAddClockInjectedOffset = False
1282 self.config.biasLevel = 0.0
1283 # Assume combined calibrations are made with 16 inputs.
1284 self.config.readNoise *= 0.25
1286 self.config.doRoundAdu = False
1289class ReferenceMockLSST(IsrMockLSST):
1290 """Parent class for those that make reference calibrations.
1291 """
1292 def __init__(self, **kwargs):
1293 # If we want the calibration in adu units, we need to apply
1294 # the gain. Default is electron units, so do not apply the gain.
1295 doApplyGain = kwargs.pop("adu", False)
1297 super().__init__(**kwargs)
1298 self.config.isTrimmed = True
1299 self.config.doGenerateImage = True
1301 self.config.calibMode = True
1303 self.config.doAddSky = False
1304 self.config.doAddSource = False
1306 self.config.doAddFringe = False
1308 self.config.doAddParallelOverscanRamp = False
1309 self.config.doAddSerialOverscanRamp = False
1310 self.config.doAddCrosstalk = False
1311 self.config.doAddBias = False
1312 self.config.doAdd2DBias = False
1313 self.config.doAddDark = False
1314 self.config.doApplyGain = doApplyGain
1315 self.config.doAddFlat = False
1316 self.config.doAddClockInjectedOffset = False
1318 # Reference calibrations are not integerized.
1319 self.config.doRoundAdu = False
1322# Classes to generate calibration products mocks.
1323class DarkMockLSST(ReferenceMockLSST):
1324 """Simulated reference dark calibration.
1325 """
1326 def __init__(self, **kwargs):
1327 super().__init__(**kwargs)
1328 self.config.doAddDark = True
1329 self.config.darkTime = 1.0
1332class BiasMockLSST(ReferenceMockLSST):
1333 """Simulated combined bias calibration.
1334 """
1335 def __init__(self, **kwargs):
1336 super().__init__(**kwargs)
1337 # This is a "2D bias residual" frame which has only
1338 # the 2D bias in it.
1339 self.config.doAdd2DBias = True
1342class FlatMockLSST(ReferenceMockLSST):
1343 """Simulated reference flat calibration.
1344 """
1345 def __init__(self, **kwargs):
1346 super().__init__(**kwargs)
1347 self.config.doAddFlat = True
1350class FringeMockLSST(ReferenceMockLSST):
1351 """Simulated reference fringe calibration.
1352 """
1353 def __init__(self, **kwargs):
1354 super().__init__(**kwargs)
1355 self.config.doAddFringe = True
1358class BfKernelMockLSST(IsrMockLSST):
1359 """Simulated brighter-fatter kernel.
1360 """
1361 def __init__(self, **kwargs):
1362 super().__init__(**kwargs)
1363 self.config.doGenerateImage = False
1364 self.config.doGenerateData = True
1366 self.config.doBrighterFatter = True
1367 self.config.doDefects = False
1368 self.config.doCrosstalkCoeffs = False
1369 self.config.doTransmissionCurve = False
1370 self.config.doLinearizer = False
1373class DeferredChargeMockLSST(IsrMockLSST):
1374 """Simulated deferred charge calibration.
1375 """
1376 def __init__(self, **kwargs):
1377 super().__init__(**kwargs)
1378 self.config.doGenerateImage = False
1379 self.config.doGenerateData = True
1380 self.config.doDeferredCharge = True
1381 self.config.doDefects = False
1382 self.config.doCrosstalkCoeffs = False
1383 self.config.doTransmissionCurve = False
1386class DefectMockLSST(IsrMockLSST):
1387 """Simulated defect list.
1388 """
1389 def __init__(self, **kwargs):
1390 super().__init__(**kwargs)
1391 self.config.doGenerateImage = False
1392 self.config.doGenerateData = True
1394 self.config.doBrighterFatter = False
1395 self.config.doDefects = True
1396 self.config.doCrosstalkCoeffs = False
1397 self.config.doTransmissionCurve = False
1398 self.config.doLinearizer = False
1401class CrosstalkCoeffMockLSST(IsrMockLSST):
1402 """Simulated crosstalk coefficient matrix.
1403 """
1404 def __init__(self, **kwargs):
1405 super().__init__(**kwargs)
1406 self.config.doGenerateImage = False
1407 self.config.doGenerateData = True
1409 self.config.doBrighterFatter = False
1410 self.config.doDefects = False
1411 self.config.doCrosstalkCoeffs = True
1412 self.config.doTransmissionCurve = False
1413 self.config.doLinearizer = False
1416class LinearizerMockLSST(IsrMockLSST):
1417 """Simulated linearizer.
1418 """
1419 def __init__(self, **kwargs):
1420 super().__init__(**kwargs)
1421 self.config.doGenerateImage = False
1422 self.config.doGenerateData = True
1424 self.config.doBrighterFatter = False
1425 self.config.doDefects = False
1426 self.config.doCrosstalkCoeffs = False
1427 self.config.doTransmissionCurve = False
1428 self.config.doLinearizer = True
1431class TransmissionMockLSST(IsrMockLSST):
1432 """Simulated transmission curve.
1433 """
1434 def __init__(self, **kwargs):
1435 super().__init__(**kwargs)
1436 self.config.doGenerateImage = False
1437 self.config.doGenerateData = True
1439 self.config.doBrighterFatter = False
1440 self.config.doDefects = False
1441 self.config.doCrosstalkCoeffs = False
1442 self.config.doTransmissionCurve = True
1443 self.config.doLinearizer = False