AstrometryTask#

class lsst.meas.astrom.AstrometryTask(refObjLoader=None, schema=None, **kwargs)#

Bases: RefMatchTask

Match an input source catalog with objects from a reference catalog and solve for the WCS.

This task is broken into two main subasks: matching and WCS fitting which are very interactive. The matching here can be considered in part a first pass WCS fitter due to the fitter’s sensitivity to outliers.

Parameters#

refObjLoaderlsst.meas.algorithms.ReferenceLoader

A reference object loader object; gen3 pipeline tasks will pass None and call setRefObjLoader in runQuantum.

schemalsst.afw.table.Schema

Used to set “calib_astrometry_used” flag in output source catalog.

**kwargs

Additional keyword arguments for pipe_base lsst.pipe.base.Task.__init__.

Methods Summary

check(exposure, sourceCat, nMatches)

Validate the astrometric fit against the maxMeanDistance threshold.

run(sourceCat, exposure)

Load reference objects, match sources and optionally fit a WCS.

solve(exposure, sourceCat)

Load reference objects overlapping an exposure, match to sources and fit a WCS

Methods Documentation

check(exposure, sourceCat, nMatches)#

Validate the astrometric fit against the maxMeanDistance threshold.

If the distMean metric does not satisfy the requirement of being less than the value set in config.maxMeanDistanceArcsec, the WCS on the exposure will be set to None and the coordinate values in the source catalog will be set to NaN to reflect a failed astrometric fit.

Parameters#

exposurelsst.afw.image.Exposure

The exposure whose astrometric fit is being evaluated.

sourceCatlsst.afw.table.SourceCatalog

The catalog of sources associated with the exposure.

nMatchesint

The number of matches that were found and used during the astrometric fit (for logging purposes only).

Raises#

BadAstrometryFit

If the measured mean on-sky distance between the matched source and reference objects is greater than self.config.maxMeanDistanceArcsec.

run(sourceCat, exposure)#

Load reference objects, match sources and optionally fit a WCS.

This is a thin layer around solve or loadAndMatch, depending on config.forceKnownWcs.

Parameters#

exposurelsst.afw.image.Exposure

exposure whose WCS is to be fit The following are read only:

  • bbox

  • filter (may be unset)

  • detector (if wcs is pure tangent; may be absent)

The following are updated:

  • wcs (the initial value is used as an initial guess, and is required)

sourceCatlsst.afw.table.SourceCatalog

catalog of sources detected on the exposure

Returns#

resultlsst.pipe.base.Struct

with these fields:

  • refCat : reference object catalog of objects that overlap the exposure (with some margin) (lsst.afw.table.SimpleCatalog).

  • matches : astrometric matches (list of lsst.afw.table.ReferenceMatch).

  • scatterOnSky : median on-sky separation between reference objects and sources in “matches” (lsst.afw.geom.Angle) or None if config.forceKnownWcs True

  • matchMeta : metadata needed to unpersist matches (lsst.daf.base.PropertyList)

solve(exposure, sourceCat)#

Load reference objects overlapping an exposure, match to sources and fit a WCS

Returns#

resultlsst.pipe.base.Struct

Result struct with components:

  • refCat : reference object catalog of objects that overlap the exposure (with some margin) (lsst::afw::table::SimpleCatalog).

  • matches : astrometric matches (list of lsst.afw.table.ReferenceMatch).

  • scatterOnSky : median on-sky separation between reference objects and sources in “matches” (lsst.geom.Angle)

  • matchMeta : metadata needed to unpersist matches (lsst.daf.base.PropertyList)

Raises#

BadAstrometryFit

If the measured mean on-sky distance between the matched source and reference objects is greater than self.config.maxMeanDistanceArcsec.

Notes#

ignores config.forceKnownWcs