Coverage for python/lsst/pipe/tasks/characterizeImage.py: 31%

179 statements  

« prev     ^ index     » next       coverage.py v7.5.1, created at 2024-05-07 03:40 -0700

1# This file is part of pipe_tasks. 

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/>. 

21 

22__all__ = ["CharacterizeImageConfig", "CharacterizeImageTask"] 

23 

24import numpy as np 

25 

26from lsstDebug import getDebugFrame 

27import lsst.afw.table as afwTable 

28import lsst.pex.config as pexConfig 

29import lsst.pipe.base as pipeBase 

30import lsst.daf.base as dafBase 

31import lsst.pipe.base.connectionTypes as cT 

32from lsst.afw.math import BackgroundList 

33from lsst.afw.table import SourceTable 

34from lsst.meas.algorithms import ( 

35 SubtractBackgroundTask, 

36 SourceDetectionTask, 

37 MeasureApCorrTask, 

38 MeasureApCorrError, 

39 MaskStreaksTask, 

40) 

41from lsst.meas.algorithms.installGaussianPsf import InstallGaussianPsfTask 

42from lsst.meas.astrom import displayAstrometry 

43from lsst.meas.base import ( 

44 SingleFrameMeasurementTask, 

45 ApplyApCorrTask, 

46 CatalogCalculationTask, 

47 IdGenerator, 

48 DetectorVisitIdGeneratorConfig, 

49) 

50from lsst.meas.deblender import SourceDeblendTask 

51import lsst.meas.extensions.shapeHSM # noqa: F401 needed for default shape plugin 

52from .measurePsf import MeasurePsfTask 

53from .repair import RepairTask 

54from .computeExposureSummaryStats import ComputeExposureSummaryStatsTask 

55from lsst.pex.exceptions import LengthError 

56from lsst.utils.timer import timeMethod 

57 

58 

59class CharacterizeImageConnections(pipeBase.PipelineTaskConnections, 

60 dimensions=("instrument", "visit", "detector")): 

61 exposure = cT.Input( 

62 doc="Input exposure data", 

63 name="postISRCCD", 

64 storageClass="Exposure", 

65 dimensions=["instrument", "exposure", "detector"], 

66 ) 

67 characterized = cT.Output( 

68 doc="Output characterized data.", 

69 name="icExp", 

70 storageClass="ExposureF", 

71 dimensions=["instrument", "visit", "detector"], 

72 ) 

73 sourceCat = cT.Output( 

74 doc="Output source catalog.", 

75 name="icSrc", 

76 storageClass="SourceCatalog", 

77 dimensions=["instrument", "visit", "detector"], 

78 ) 

79 backgroundModel = cT.Output( 

80 doc="Output background model.", 

81 name="icExpBackground", 

82 storageClass="Background", 

83 dimensions=["instrument", "visit", "detector"], 

84 ) 

85 outputSchema = cT.InitOutput( 

86 doc="Schema of the catalog produced by CharacterizeImage", 

87 name="icSrc_schema", 

88 storageClass="SourceCatalog", 

89 ) 

90 

91 def adjustQuantum(self, inputs, outputs, label, dataId): 

92 # Docstring inherited from PipelineTaskConnections 

93 try: 

94 return super().adjustQuantum(inputs, outputs, label, dataId) 

95 except pipeBase.ScalarError as err: 

96 raise pipeBase.ScalarError( 

97 "CharacterizeImageTask can at present only be run on visits that are associated with " 

98 "exactly one exposure. Either this is not a valid exposure for this pipeline, or the " 

99 "snap-combination step you probably want hasn't been configured to run between ISR and " 

100 "this task (as of this writing, that would be because it hasn't been implemented yet)." 

101 ) from err 

102 

103 

104class CharacterizeImageConfig(pipeBase.PipelineTaskConfig, 

105 pipelineConnections=CharacterizeImageConnections): 

106 """Config for CharacterizeImageTask.""" 

107 

108 doMeasurePsf = pexConfig.Field( 

109 dtype=bool, 

110 default=True, 

111 doc="Measure PSF? If False then for all subsequent operations use either existing PSF " 

112 "model when present, or install simple PSF model when not (see installSimplePsf " 

113 "config options)" 

114 ) 

115 doWrite = pexConfig.Field( 

116 dtype=bool, 

117 default=True, 

118 doc="Persist results?", 

119 ) 

120 doWriteExposure = pexConfig.Field( 

121 dtype=bool, 

122 default=True, 

123 doc="Write icExp and icExpBackground in addition to icSrc? Ignored if doWrite False.", 

124 ) 

125 psfIterations = pexConfig.RangeField( 

126 dtype=int, 

127 default=2, 

128 min=1, 

129 doc="Number of iterations of detect sources, measure sources, " 

130 "estimate PSF. If useSimplePsf is True then 2 should be plenty; " 

131 "otherwise more may be wanted.", 

132 ) 

133 background = pexConfig.ConfigurableField( 

134 target=SubtractBackgroundTask, 

135 doc="Configuration for initial background estimation", 

136 ) 

137 detection = pexConfig.ConfigurableField( 

138 target=SourceDetectionTask, 

139 doc="Detect sources" 

140 ) 

141 doDeblend = pexConfig.Field( 

142 dtype=bool, 

143 default=True, 

144 doc="Run deblender input exposure" 

145 ) 

146 deblend = pexConfig.ConfigurableField( 

147 target=SourceDeblendTask, 

148 doc="Split blended source into their components" 

149 ) 

150 measurement = pexConfig.ConfigurableField( 

151 target=SingleFrameMeasurementTask, 

152 doc="Measure sources" 

153 ) 

154 doApCorr = pexConfig.Field( 

155 dtype=bool, 

156 default=True, 

157 doc="Run subtasks to measure and apply aperture corrections" 

158 ) 

159 measureApCorr = pexConfig.ConfigurableField( 

160 target=MeasureApCorrTask, 

161 doc="Subtask to measure aperture corrections" 

162 ) 

163 applyApCorr = pexConfig.ConfigurableField( 

164 target=ApplyApCorrTask, 

165 doc="Subtask to apply aperture corrections" 

166 ) 

167 # If doApCorr is False, and the exposure does not have apcorrections already applied, the 

168 # active plugins in catalogCalculation almost certainly should not contain the characterization plugin 

169 catalogCalculation = pexConfig.ConfigurableField( 

170 target=CatalogCalculationTask, 

171 doc="Subtask to run catalogCalculation plugins on catalog" 

172 ) 

173 doComputeSummaryStats = pexConfig.Field( 

174 dtype=bool, 

175 default=True, 

176 doc="Run subtask to measure exposure summary statistics", 

177 deprecated=("This subtask has been moved to CalibrateTask " 

178 "with DM-30701.") 

179 ) 

180 computeSummaryStats = pexConfig.ConfigurableField( 

181 target=ComputeExposureSummaryStatsTask, 

182 doc="Subtask to run computeSummaryStats on exposure", 

183 deprecated=("This subtask has been moved to CalibrateTask " 

184 "with DM-30701.") 

185 ) 

186 useSimplePsf = pexConfig.Field( 

187 dtype=bool, 

188 default=True, 

189 doc="Replace the existing PSF model with a simplified version that has the same sigma " 

190 "at the start of each PSF determination iteration? Doing so makes PSF determination " 

191 "converge more robustly and quickly.", 

192 ) 

193 installSimplePsf = pexConfig.ConfigurableField( 

194 target=InstallGaussianPsfTask, 

195 doc="Install a simple PSF model", 

196 ) 

197 measurePsf = pexConfig.ConfigurableField( 

198 target=MeasurePsfTask, 

199 doc="Measure PSF", 

200 ) 

201 repair = pexConfig.ConfigurableField( 

202 target=RepairTask, 

203 doc="Remove cosmic rays", 

204 ) 

205 requireCrForPsf = pexConfig.Field( 

206 dtype=bool, 

207 default=True, 

208 doc="Require cosmic ray detection and masking to run successfully before measuring the PSF." 

209 ) 

210 checkUnitsParseStrict = pexConfig.Field( 

211 doc="Strictness of Astropy unit compatibility check, can be 'raise', 'warn' or 'silent'", 

212 dtype=str, 

213 default="raise", 

214 ) 

215 doMaskStreaks = pexConfig.Field( 

216 doc="Mask streaks", 

217 default=True, 

218 dtype=bool, 

219 ) 

220 maskStreaks = pexConfig.ConfigurableField( 

221 target=MaskStreaksTask, 

222 doc="Subtask for masking streaks. Only used if doMaskStreaks is True. " 

223 "Adds a mask plane to an exposure, with the mask plane name set by streakMaskName.", 

224 ) 

225 idGenerator = DetectorVisitIdGeneratorConfig.make_field() 

226 

227 def setDefaults(self): 

228 super().setDefaults() 

229 # just detect bright stars; includeThresholdMultipler=10 seems large, 

230 # but these are the values we have been using 

231 self.detection.thresholdValue = 5.0 

232 self.detection.includeThresholdMultiplier = 10.0 

233 # do not deblend, as it makes a mess 

234 self.doDeblend = False 

235 # measure and apply aperture correction; note: measuring and applying aperture 

236 # correction are disabled until the final measurement, after PSF is measured 

237 self.doApCorr = True 

238 # During characterization, we don't have full source measurement information, 

239 # so must do the aperture correction with only psf stars, combined with the 

240 # default signal-to-noise cuts in MeasureApCorrTask. 

241 selector = self.measureApCorr.sourceSelector["science"] 

242 selector.doUnresolved = False 

243 selector.flags.good = ["calib_psf_used"] 

244 selector.flags.bad = [] 

245 

246 # minimal set of measurements needed to determine PSF 

247 self.measurement.plugins.names = [ 

248 "base_PixelFlags", 

249 "base_SdssCentroid", 

250 "ext_shapeHSM_HsmSourceMoments", 

251 "base_GaussianFlux", 

252 "base_PsfFlux", 

253 "base_CircularApertureFlux", 

254 "base_ClassificationSizeExtendedness", 

255 ] 

256 self.measurement.slots.shape = "ext_shapeHSM_HsmSourceMoments" 

257 

258 def validate(self): 

259 if self.doApCorr and not self.measurePsf: 

260 raise RuntimeError("Must measure PSF to measure aperture correction, " 

261 "because flags determined by PSF measurement are used to identify " 

262 "sources used to measure aperture correction") 

263 

264 

265class CharacterizeImageTask(pipeBase.PipelineTask): 

266 """Measure bright sources and use this to estimate background and PSF of 

267 an exposure. 

268 

269 Given an exposure with defects repaired (masked and interpolated over, 

270 e.g. as output by `~lsst.ip.isr.IsrTask`): 

271 - detect and measure bright sources 

272 - repair cosmic rays 

273 - detect and mask streaks 

274 - measure and subtract background 

275 - measure PSF 

276 

277 Parameters 

278 ---------- 

279 schema : `lsst.afw.table.Schema`, optional 

280 Initial schema for icSrc catalog. 

281 **kwargs 

282 Additional keyword arguments. 

283 

284 Notes 

285 ----- 

286 Debugging: 

287 CharacterizeImageTask has a debug dictionary with the following keys: 

288 

289 frame 

290 int: if specified, the frame of first debug image displayed (defaults to 1) 

291 repair_iter 

292 bool; if True display image after each repair in the measure PSF loop 

293 background_iter 

294 bool; if True display image after each background subtraction in the measure PSF loop 

295 measure_iter 

296 bool; if True display image and sources at the end of each iteration of the measure PSF loop 

297 See `~lsst.meas.astrom.displayAstrometry` for the meaning of the various symbols. 

298 psf 

299 bool; if True display image and sources after PSF is measured; 

300 this will be identical to the final image displayed by measure_iter if measure_iter is true 

301 repair 

302 bool; if True display image and sources after final repair 

303 measure 

304 bool; if True display image and sources after final measurement 

305 """ 

306 

307 ConfigClass = CharacterizeImageConfig 

308 _DefaultName = "characterizeImage" 

309 

310 def __init__(self, schema=None, **kwargs): 

311 super().__init__(**kwargs) 

312 

313 if schema is None: 

314 schema = SourceTable.makeMinimalSchema() 

315 self.schema = schema 

316 self.makeSubtask("background") 

317 self.makeSubtask("installSimplePsf") 

318 self.makeSubtask("repair") 

319 if self.config.doMaskStreaks: 

320 self.makeSubtask("maskStreaks") 

321 self.makeSubtask("measurePsf", schema=self.schema) 

322 self.algMetadata = dafBase.PropertyList() 

323 self.makeSubtask('detection', schema=self.schema) 

324 if self.config.doDeblend: 

325 self.makeSubtask("deblend", schema=self.schema) 

326 self.makeSubtask('measurement', schema=self.schema, algMetadata=self.algMetadata) 

327 if self.config.doApCorr: 

328 self.makeSubtask('measureApCorr', schema=self.schema) 

329 self.makeSubtask('applyApCorr', schema=self.schema) 

330 self.makeSubtask('catalogCalculation', schema=self.schema) 

331 self._initialFrame = getDebugFrame(self._display, "frame") or 1 

332 self._frame = self._initialFrame 

333 self.schema.checkUnits(parse_strict=self.config.checkUnitsParseStrict) 

334 self.outputSchema = afwTable.SourceCatalog(self.schema) 

335 

336 def runQuantum(self, butlerQC, inputRefs, outputRefs): 

337 inputs = butlerQC.get(inputRefs) 

338 if 'idGenerator' not in inputs.keys(): 

339 inputs['idGenerator'] = self.config.idGenerator.apply(butlerQC.quantum.dataId) 

340 outputs = self.run(**inputs) 

341 butlerQC.put(outputs, outputRefs) 

342 

343 @timeMethod 

344 def run(self, exposure, background=None, idGenerator=None): 

345 """Characterize a science image. 

346 

347 Peforms the following operations: 

348 - Iterate the following config.psfIterations times, or once if config.doMeasurePsf false: 

349 - detect and measure sources and estimate PSF (see detectMeasureAndEstimatePsf for details) 

350 - interpolate over cosmic rays 

351 - perform final measurement 

352 

353 Parameters 

354 ---------- 

355 exposure : `lsst.afw.image.ExposureF` 

356 Exposure to characterize. 

357 background : `lsst.afw.math.BackgroundList`, optional 

358 Initial model of background already subtracted from exposure. 

359 idGenerator : `lsst.meas.base.IdGenerator`, optional 

360 Object that generates source IDs and provides RNG seeds. 

361 

362 Returns 

363 ------- 

364 result : `lsst.pipe.base.Struct` 

365 Results as a struct with attributes: 

366 

367 ``exposure`` 

368 Characterized exposure (`lsst.afw.image.ExposureF`). 

369 ``sourceCat`` 

370 Detected sources (`lsst.afw.table.SourceCatalog`). 

371 ``background`` 

372 Model of subtracted background (`lsst.afw.math.BackgroundList`). 

373 ``psfCellSet`` 

374 Spatial cells of PSF candidates (`lsst.afw.math.SpatialCellSet`). 

375 ``characterized`` 

376 Another reference to ``exposure`` for compatibility. 

377 ``backgroundModel`` 

378 Another reference to ``background`` for compatibility. 

379 

380 Raises 

381 ------ 

382 RuntimeError 

383 Raised if PSF sigma is NaN. 

384 """ 

385 self._frame = self._initialFrame # reset debug display frame 

386 

387 if not self.config.doMeasurePsf and not exposure.hasPsf(): 

388 self.log.info("CharacterizeImageTask initialized with 'simple' PSF.") 

389 self.installSimplePsf.run(exposure=exposure) 

390 

391 if idGenerator is None: 

392 idGenerator = IdGenerator() 

393 

394 # subtract an initial estimate of background level 

395 background = self.background.run(exposure).background 

396 

397 psfIterations = self.config.psfIterations if self.config.doMeasurePsf else 1 

398 for i in range(psfIterations): 

399 dmeRes = self.detectMeasureAndEstimatePsf( 

400 exposure=exposure, 

401 idGenerator=idGenerator, 

402 background=background, 

403 ) 

404 

405 psf = dmeRes.exposure.getPsf() 

406 # Just need a rough estimate; average positions are fine 

407 psfAvgPos = psf.getAveragePosition() 

408 psfSigma = psf.computeShape(psfAvgPos).getDeterminantRadius() 

409 psfDimensions = psf.computeImage(psfAvgPos).getDimensions() 

410 medBackground = np.median(dmeRes.background.getImage().getArray()) 

411 self.log.info("iter %s; PSF sigma=%0.4f, dimensions=%s; median background=%0.2f", 

412 i + 1, psfSigma, psfDimensions, medBackground) 

413 if np.isnan(psfSigma): 

414 raise RuntimeError("PSF sigma is NaN, cannot continue PSF determination.") 

415 

416 self.display("psf", exposure=dmeRes.exposure, sourceCat=dmeRes.sourceCat) 

417 

418 # perform final repair with final PSF 

419 self.repair.run(exposure=dmeRes.exposure) 

420 self.display("repair", exposure=dmeRes.exposure, sourceCat=dmeRes.sourceCat) 

421 

422 # mask streaks 

423 if self.config.doMaskStreaks: 

424 _ = self.maskStreaks.run(dmeRes.exposure) 

425 

426 # perform final measurement with final PSF, including measuring and applying aperture correction, 

427 # if wanted 

428 self.measurement.run(measCat=dmeRes.sourceCat, exposure=dmeRes.exposure, 

429 exposureId=idGenerator.catalog_id) 

430 if self.config.doApCorr: 

431 try: 

432 apCorrMap = self.measureApCorr.run( 

433 exposure=dmeRes.exposure, 

434 catalog=dmeRes.sourceCat, 

435 ).apCorrMap 

436 except MeasureApCorrError: 

437 # We have failed to get a valid aperture correction map. 

438 # Proceed with processing, and image will be filtered 

439 # downstream. 

440 dmeRes.exposure.info.setApCorrMap(None) 

441 else: 

442 dmeRes.exposure.info.setApCorrMap(apCorrMap) 

443 self.applyApCorr.run(catalog=dmeRes.sourceCat, apCorrMap=exposure.getInfo().getApCorrMap()) 

444 

445 self.catalogCalculation.run(dmeRes.sourceCat) 

446 

447 self.display("measure", exposure=dmeRes.exposure, sourceCat=dmeRes.sourceCat) 

448 

449 return pipeBase.Struct( 

450 exposure=dmeRes.exposure, 

451 sourceCat=dmeRes.sourceCat, 

452 background=dmeRes.background, 

453 psfCellSet=dmeRes.psfCellSet, 

454 

455 characterized=dmeRes.exposure, 

456 backgroundModel=dmeRes.background 

457 ) 

458 

459 @timeMethod 

460 def detectMeasureAndEstimatePsf(self, exposure, idGenerator, background): 

461 """Perform one iteration of detect, measure, and estimate PSF. 

462 

463 Performs the following operations: 

464 

465 - if config.doMeasurePsf or not exposure.hasPsf(): 

466 

467 - install a simple PSF model (replacing the existing one, if need be) 

468 

469 - interpolate over cosmic rays with keepCRs=True 

470 - estimate background and subtract it from the exposure 

471 - detect, deblend and measure sources, and subtract a refined background model; 

472 - if config.doMeasurePsf: 

473 - measure PSF 

474 

475 Parameters 

476 ---------- 

477 exposure : `lsst.afw.image.ExposureF` 

478 Exposure to characterize. 

479 idGenerator : `lsst.meas.base.IdGenerator` 

480 Object that generates source IDs and provides RNG seeds. 

481 background : `lsst.afw.math.BackgroundList`, optional 

482 Initial model of background already subtracted from exposure. 

483 

484 Returns 

485 ------- 

486 result : `lsst.pipe.base.Struct` 

487 Results as a struct with attributes: 

488 

489 ``exposure`` 

490 Characterized exposure (`lsst.afw.image.ExposureF`). 

491 ``sourceCat`` 

492 Detected sources (`lsst.afw.table.SourceCatalog`). 

493 ``background`` 

494 Model of subtracted background (`lsst.afw.math.BackgroundList`). 

495 ``psfCellSet`` 

496 Spatial cells of PSF candidates (`lsst.afw.math.SpatialCellSet`). 

497 

498 Raises 

499 ------ 

500 LengthError 

501 Raised if there are too many CR pixels. 

502 """ 

503 # install a simple PSF model, if needed or wanted 

504 if not exposure.hasPsf() or (self.config.doMeasurePsf and self.config.useSimplePsf): 

505 self.log.info("PSF estimation initialized with 'simple' PSF") 

506 self.installSimplePsf.run(exposure=exposure) 

507 

508 # run repair, but do not interpolate over cosmic rays (do that elsewhere, with the final PSF model) 

509 if self.config.requireCrForPsf: 

510 self.repair.run(exposure=exposure, keepCRs=True) 

511 else: 

512 try: 

513 self.repair.run(exposure=exposure, keepCRs=True) 

514 except LengthError: 

515 self.log.warning("Skipping cosmic ray detection: Too many CR pixels (max %0.f)", 

516 self.config.repair.cosmicray.nCrPixelMax) 

517 

518 self.display("repair_iter", exposure=exposure) 

519 

520 if background is None: 

521 background = BackgroundList() 

522 

523 sourceIdFactory = idGenerator.make_table_id_factory() 

524 table = SourceTable.make(self.schema, sourceIdFactory) 

525 table.setMetadata(self.algMetadata) 

526 

527 detRes = self.detection.run(table=table, exposure=exposure, doSmooth=True) 

528 sourceCat = detRes.sources 

529 if detRes.background: 

530 for bg in detRes.background: 

531 background.append(bg) 

532 

533 if self.config.doDeblend: 

534 self.deblend.run(exposure=exposure, sources=sourceCat) 

535 # We need the output catalog to be contiguous for further processing. 

536 if not sourceCat.isContiguous(): 

537 sourceCat = sourceCat.copy(deep=True) 

538 

539 self.measurement.run(measCat=sourceCat, exposure=exposure, exposureId=idGenerator.catalog_id) 

540 

541 measPsfRes = pipeBase.Struct(cellSet=None) 

542 if self.config.doMeasurePsf: 

543 measPsfRes = self.measurePsf.run(exposure=exposure, sources=sourceCat, 

544 expId=idGenerator.catalog_id) 

545 self.display("measure_iter", exposure=exposure, sourceCat=sourceCat) 

546 

547 return pipeBase.Struct( 

548 exposure=exposure, 

549 sourceCat=sourceCat, 

550 background=background, 

551 psfCellSet=measPsfRes.cellSet, 

552 ) 

553 

554 def display(self, itemName, exposure, sourceCat=None): 

555 """Display exposure and sources on next frame (for debugging). 

556 

557 Parameters 

558 ---------- 

559 itemName : `str` 

560 Name of item in ``debugInfo``. 

561 exposure : `lsst.afw.image.ExposureF` 

562 Exposure to display. 

563 sourceCat : `lsst.afw.table.SourceCatalog`, optional 

564 Catalog of sources detected on the exposure. 

565 """ 

566 val = getDebugFrame(self._display, itemName) 

567 if not val: 

568 return 

569 

570 displayAstrometry(exposure=exposure, sourceCat=sourceCat, frame=self._frame, pause=False) 

571 self._frame += 1