def plotHist(exposure, name="", numOfBin=1000, log=False, saveFilePath=None):
"""
Plot the histogram.
Arguments:
exposure {[exposure]} -- Data butler of exposure image.
Keyword Arguments:
name {string} -- Image title name (default: {""}).
numOfBin {int} -- Number of bins (default: {1000}).
log {bool} -- The histogram axis will be set to a log scale if log=True
(default: {False}).
saveFilePath {[str]} -- Save image to file path. (default: {None})
"""
# Get the image data
img = ButlerWrapper.getImageData(exposure)
# Plot the histogram
plt.figure()
plt.hist(img.flatten(), bins=int(numOfBin), log=log)
plt.title(name)
if (saveFilePath is not None):
plt.savefig(saveFilePath, bbox_inches="tight")
plt.close()
else:
plt.show()
def testGetImageData(self):
eimg = self._getEimage()
image = ButlerWrapper.getImageData(eimg)
self.assertTrue(isinstance(image, np.ndarray))
self.assertEqual(image.shape, (4000, 4072))
def testGetImageData(self):
exposure = self._getRawExp()
image = ButlerWrapper.getImageData(exposure)
self.assertTrue(isinstance(image, np.ndarray))
self.assertEqual(image.shape, (4020, 4176))
def plotHist(exposure, name="", numOfBin=1000, log=False, saveFilePath=None):
"""Plot the histogram.
Parameters
----------
exposure : Exposure
Data butler of exposure image.
name : str, optional
Image title name. (the default is "".)
numOfBin : int, optional
Number of bins. (the default is 1000.)
log : bool, optional
The histogram axis will be set to a log scale if log=True. (the default
is False.)
saveFilePath : str, optional
Save image to file path. (the default is None.)
"""
# Get the image data
img = ButlerWrapper.getImageData(exposure)
# Plot the histogram
plt.figure()
plt.hist(img.flatten(), bins=int(numOfBin), log=log)
plt.title(name)
if saveFilePath is None:
plt.show()
else:
plt.savefig(saveFilePath, bbox_inches="tight")
plt.close()
def plotHist(exposure, name="", numOfBin=1000, log=False, saveFilePath=None):
"""Plot the histogram.
Parameters
----------
exposure : Exposure
Data butler of exposure image.
name : str, optional
Image title name. (the default is "".)
numOfBin : int, optional
Number of bins. (the default is 1000.)
log : bool, optional
The histogram axis will be set to a log scale if log=True. (the default
is False.)
saveFilePath : str, optional
Save image to file path. (the default is None.)
"""
# Get the image data
img = ButlerWrapper.getImageData(exposure)
# Plot the histogram
plt.figure()
plt.hist(img.flatten(), bins=int(numOfBin), log=log)
plt.title(name)
if (saveFilePath is not None):
plt.savefig(saveFilePath, bbox_inches="tight")
plt.close()
else:
plt.show()
def poltExposureImage(exposure, name="", scale="log", cmap="gray", vmin=None,
vmax=None, saveFilePath=None):
"""Plot the exposure image.
Parameters
----------
exposure : Exposure
Data butler of exposure image.
name : str, optional
Image title name. (the default is "".)
scale : str, optional
Scale of image map (log or linear). (the default is "log".)
cmap : str, optional
Color map definition. (the default is "gray".)
vmin : float, optional
Mininum value to show. This normalizes the luminance data. (the default
is None.)
vmax : float, optional
Maximum value to show. This normalizes the luminance data. (the default
is None.)
saveFilePath : str, optional
Save image to file path. (the default is None.)
"""
# Get the image data
img = ButlerWrapper.getImageData(exposure)
# Change the scale if needed
if scale not in ("linear", "log"):
print("No %s scale to choose. Only 'linear' and 'log' scales are allowed." % scale)
return
# Decide the norm in imshow for the ploting
if (scale == "linear"):
plotNorm = None
elif (scale == "log"):
if (img.min()) < 0:
plotNorm = SymLogNorm(linthresh=0.03)
else:
plotNorm = LogNorm()
# Plot the image
plt.figure()
plt.imshow(img, cmap=cmap, origin="lower", norm=plotNorm, vmin=vmin,
vmax=vmax)
plt.colorbar()
plt.title(name)
if (saveFilePath is not None):
plt.savefig(saveFilePath, bbox_inches="tight")
plt.close()
else:
plt.show()
def poltExposureImage(exposure, name="", scale="log", cmap="gray", vmin=None, vmax=None, saveFilePath=None):
"""
Plot the exposure image.
Arguments:
exposure {[exposure]} -- Data butler of exposure image.
Keyword Arguments:
name {[str]} -- Image title name. (default: {""})
scale {[str]} -- Scale of image map (log or linear). (default: {"log"})
cmap {[str]} -- Color map definition. (default: {"gray"})
vmin {[float]} -- Mininum value to show. This normalizes the luminance data. (default: {None})
vmax {[float]} -- Maximum value to show. This normalizes the luminance data. (default: {None})
saveFilePath {[str]} -- Save image to file path. (default: {None})
"""
# Get the image data
img = ButlerWrapper.getImageData(exposure)
# Change the scale if needed
if scale not in ("linear", "log"):
print("No %s scale to choose. Only 'linear' and 'log' scales are allowed." % scale)
return
# Decide the norm in imshow for the ploting
if (scale == "linear"):
plotNorm = None
elif (scale == "log"):
if (img.min()) < 0:
plotNorm = SymLogNorm(linthresh=0.03)
else:
plotNorm = LogNorm()
# Plot the image
plt.figure()
plt.imshow(img, cmap=cmap, origin="lower", norm=plotNorm, vmin=vmin, vmax=vmax)
plt.colorbar()
plt.title(name)
if (saveFilePath is not None):
plt.savefig(saveFilePath, bbox_inches="tight")
plt.close()
else:
plt.show()
class WepController(object):
CORNER_WFS_LIST = [
"R00_SW0",
"R00_SW1",
"R04_SW0",
"R04_SW1",
"R40_SW0",
"R40_SW1",
"R44_SW0",
"R44_SW1",
]
def __init__(self, dataCollector, isrWrapper, sourSelc, sourProc, wfEsti):
"""Initialize the wavefront estimation pipeline (WEP) controller class.
Parameters
----------
dataCollector : CamDataCollector
Camera data collector.
isrWrapper : CamIsrWrapper
Instrument signature removal (ISR) wrapper.
sourSelc : SourceSelector
Source selector.
sourProc : SourceProcessor
Source processor.
wfEsti : WfEstimator
Wavefront estimator.
"""
# Data collector to use DM ingest command line task
self.dataCollector = dataCollector
# ISR wrapper to use DM ISR command line task
self.isrWrapper = isrWrapper
# Source selector to query the bright star catalog
self.sourSelc = sourSelc
# Source processor to get the donut image or do the deblending
self.sourProc = sourProc
# Wavefront estimator to get zk
self.wfEsti = wfEsti
# Butler wrapper to use DM data butler
self.butlerWrapper = None
def getDataCollector(self):
"""Get the attribute of data collector.
Returns
-------
CamDataCollector
Data collector.
"""
return self.dataCollector
def getIsrWrapper(self):
"""Get the attribute of ISR wraper.
ISR: Instrument signature removal.
Returns
-------
CamIsrWrapper
ISR wraper.
"""
return self.isrWrapper
def getSourSelc(self):
"""Get the attribute of source selector.
Returns
-------
SourceSelector
Source selector.
"""
return self.sourSelc
def getSourProc(self):
"""Get the attribute of source processor.
Returns
-------
SourceProcessor
Source processor.
"""
return self.sourProc
def getWfEsti(self):
"""Get the attribute of wavefront estimator.
Returns
-------
WfEstimator
Wavefront estimator.
"""
return self.wfEsti
def getButlerWrapper(self):
"""Get the attribute of butler wrapper.
Returns
-------
ButlerWrapper
Butler wrapper.
"""
return self.butlerWrapper
def setPostIsrCcdInputs(self, inputs):
"""Set inputs of post instrument signature removal (ISR) CCD images.
Parameters
----------
inputs : RepositoryArgs, dict, or str
Can be a single item or a list. Provides arguments to load an
existing repository (or repositories). String is assumed to be a
URI and is used as the cfgRoot (URI to the location of the cfg
file). (Local file system URI does not have to start with
'file://' and in this way can be a relative path). The
'RepositoryArgs' class can be used to provide more parameters with
which to initialize a repository (such as 'mapper', 'mapperArgs',
'tags', etc. See the 'RepositoryArgs' documentation for more
details). A dict may be used as shorthand for a 'RepositoryArgs'
class instance. The dict keys must match parameters to the
'RepositoryArgs.__init__' function.
"""
self.butlerWrapper = ButlerWrapper(inputs)
def getPostIsrImgMapByPistonDefocal(self, sensorNameList, obsIdList):
"""Get the post ISR image map that the defocal images are by the
pistion motion.
Parameters
----------
sensorNameList : list
List of sensor name.
obsIdList : list
Observation Id list in [intraObsId, extraObsId].
Returns
-------
dict
Post-ISR image map. The dictionary key is the sensor name. The
dictionary item is the defocal image on the camera coordinate.
(type: DefocalImage).
"""
return self._getImgMapByPistonDefocal(sensorNameList, obsIdList, ImageType.Amp)
def _getImgMapByPistonDefocal(self, sensorNameList, obsIdList, imageType):
"""Get the image map that the defocal images are by the pistion motion.
Parameters
----------
sensorNameList : list
List of sensor name.
obsIdList : list
Observation Id list in [intraObsId, extraObsId].
imageType : ImageType
Image type.
Returns
-------
dict
Image map. The dictionary key is the sensor name. The dictionary
item is the defocal image on the camera coordinate. (type:
DefocalImage).
Raises
------
ValueError
The image type is not supported.
"""
# Get the waveront image map
wfsImgMap = dict()
for sensorName in sensorNameList:
# Get the sensor name information
raft, sensor = self._getSensorInfo(sensorName)
# The intra/ extra defocal images are decided by obsId
imgList = []
for visit in obsIdList:
# Get the exposure image in ndarray
if imageType == ImageType.Amp:
exp = self.butlerWrapper.getPostIsrCcd(int(visit), raft, sensor)
elif imageType == ImageType.Eimg:
exp = self.butlerWrapper.getEimage(int(visit), raft, sensor)
else:
raise ValueError("The %s is not supported." % imageType)
img = self.butlerWrapper.getImageData(exp)
# Transform the image in DM coordinate to camera coordinate.
camImg = self._transImgDmCoorToCamCoor(img)
# Collect the image
imgList.append(camImg)
wfsImgMap[sensorName] = DefocalImage(
intraImg=imgList[0], extraImg=imgList[1]
)
return wfsImgMap
def _getSensorInfo(self, sensorName):
"""Get the sensor information.
Parameters
----------
sensorName : str
Sensor name (e.g. "R22_S11" or "R44_SW0"). Note: SW0 and SW1 are
used as the wavefront sensor in lsst.obs.lsst.LsstCamMapper.
Returns
-------
str
Raft.
str
Sensor.
Raises
------
ValueError
Can not match the raft and sensor from sensorName.
"""
m = re.match(r"R(\d\d)_S([W\d]\d)", sensorName)
if m is not None:
raftId, sensorId = m.groups()
raft = "R" + raftId
sensor = "S" + sensorId
return raft, sensor
else:
raise ValueError(f"Can not match the raft and sensor from {sensorName}")
def _transImgDmCoorToCamCoor(self, dmImg):
"""Transfrom the image in DM coordinate to camera coordinate.
The geometry transformation is defined in LSE-349.
Parameters
----------
dmImg : numpy.ndarray
Image in DM coordinate.
Returns
-------
numpy.ndarray
Image is camera coordinate.
"""
# For the PhoSim mapper, it is the transpose (x, y) to (y, x).
camImg = dmImg.T
return camImg
def getEimgMapByPistonDefocal(self, sensorNameList, obsIdList):
"""Get the eimage map that the defocal images are by the pistion
motion.
Parameters
----------
sensorNameList : list
List of sensor name.
obsIdList : list
Observation Id list in [intraObsId, extraObsId].
Returns
-------
dict
Eimage map. The dictionary key is the sensor name. The dictionary
item is the defocal image on the camera coordinate. (type:
DefocalImage).
"""
return self._getImgMapByPistonDefocal(sensorNameList, obsIdList, ImageType.Eimg)
def getPostIsrImgMapOnCornerWfs(self, sensorNameList, obsId):
"""Get the post ISR image map of corner wavefront sensors.
Parameters
----------
sensorNameList : list
List of sensor name.
obsId : int
Observation Id.
Returns
-------
dict
Post-ISR image map. The dictionary key is the sensor name. The
dictionary item is the defocal image on the camera coordinate.
(type: DefocalImage).
"""
pass
def getDonutMap(self, neighborStarMap, wfsImgMap, filterType, doDeblending=False):
"""Get the donut map on each wavefront sensor (WFS).
Parameters
----------
neighborStarMap : dict
Information of neighboring stars and candidate stars with the name
of sensor as a dictionary.
wfsImgMap : dict
Post-ISR image map. The dictionary key is the sensor name. The
dictionary item is the defocal image on the camera coordinate.
(type: DefocalImage).
filterType : FilterType
Filter type.
doDeblending : bool, optional
Do the deblending or not. If False, only consider the single donut
based on the bright star catalog.(the default is False.)
Returns
-------
dict
Donut image map. The dictionary key is the sensor name. The
dictionary item is the list of donut image (type:
list[DonutImage]).
"""
donutMap = dict()
for sensorName, nbrStar in neighborStarMap.items():
# Configure the source processor
self.sourProc.config(sensorName=sensorName)
# Get the defocal images: [intra, extra]
defocalImgList = [
wfsImgMap[sensorName].getIntraImg(),
wfsImgMap[sensorName].getExtraImg(),
]
# Get the bright star id list on specific sensor
brightStarIdList = list(nbrStar.getId())
for starIdIdx in range(len(brightStarIdList)):
# Get the single star map
for jj in range(len(defocalImgList)):
ccdImg = defocalImgList[jj]
# Get the segment of image
if ccdImg is not None:
(
singleSciNeiImg,
allStarPosX,
allStarPosY,
magRatio,
offsetX,
offsetY,
) = self.sourProc.getSingleTargetImage(
ccdImg, nbrStar, starIdIdx, filterType
)
# Only consider the single donut if no deblending
if (not doDeblending) and (len(magRatio) != 1):
continue
# Get the single donut/ deblended image
if (len(magRatio) == 1) or (not doDeblending):
imgDeblend = singleSciNeiImg
if len(magRatio) == 1:
realcx, realcy = searchDonutPos(imgDeblend)
else:
realcx = allStarPosX[-1]
realcy = allStarPosY[-1]
# Do the deblending or not
elif len(magRatio) == 2 and doDeblending:
imgDeblend, realcx, realcy = self.sourProc.doDeblending(
singleSciNeiImg, allStarPosX, allStarPosY, magRatio
)
# Update the magnitude ratio
magRatio = [1]
else:
continue
# Extract the image
if len(magRatio) == 1:
sizeInPix = self.wfEsti.getSizeInPix()
x0 = np.floor(realcx - sizeInPix / 2).astype("int")
y0 = np.floor(realcy - sizeInPix / 2).astype("int")
imgDeblend = imgDeblend[
y0 : y0 + sizeInPix, x0 : x0 + sizeInPix
]
# Rotate the image if the sensor is the corner
# wavefront sensor
if sensorName in self.CORNER_WFS_LIST:
# Get the Euler angle
eulerZangle = round(
self.sourProc.getEulerZinDeg(sensorName)
)
# Change the sign if the angle < 0
while eulerZangle < 0:
eulerZangle += 360
# Do the rotation of matrix
numOfRot90 = eulerZangle // 90
imgDeblend = np.flipud(
np.rot90(np.flipud(imgDeblend), numOfRot90)
)
# Put the deblended image into the donut map
if sensorName not in donutMap.keys():
donutMap[sensorName] = []
# Check the donut exists in the list or not
starId = brightStarIdList[starIdIdx]
donutIndex = self._searchDonutListId(
donutMap[sensorName], starId
)
# Create the donut object and put into the list if it
# is needed
if donutIndex < 0:
# Calculate the field X, Y
pixelX = realcx + offsetX
pixelY = realcy + offsetY
fieldX, fieldY = self.sourProc.camXYtoFieldXY(
pixelX, pixelY
)
# Instantiate the DonutImage class
donutImg = DonutImage(
starId, pixelX, pixelY, fieldX, fieldY
)
donutMap[sensorName].append(donutImg)
# Search for the donut index again
donutIndex = self._searchDonutListId(
donutMap[sensorName], starId
)
# Get the donut image list
donutList = donutMap[sensorName]
# Take the absolute value for images, which might
# contain the negative value after the ISR correction.
# This happens for the amplifier images.
imgDeblend = np.abs(imgDeblend)
# Set the intra focal image
if jj == 0:
donutList[donutIndex].setImg(intraImg=imgDeblend)
# Set the extra focal image
elif jj == 1:
donutList[donutIndex].setImg(extraImg=imgDeblend)
return donutMap
def _searchDonutListId(self, donutList, starId):
"""Search the bright star ID in the donut list.
Parameters
----------
donutList : list
List of DonutImage object.
starId : int
Star Id.
Returns
-------
int
Index of donut image object with the specific starId.
"""
index = -1
for ii in range(len(donutList)):
if donutList[ii].getStarId() == int(starId):
index = ii
break
return index
def calcWfErr(self, donutMap):
"""Calculate the wavefront error in annular Zernike polynomials
(z4-z22).
Parameters
----------
donutMap : dict
Donut image map. The dictionary key is the sensor name. The
dictionary item is the donut image (type: DonutImage).
Returns
-------
dict
Donut image map with calculated wavefront error.
"""
for sensorName, donutList in donutMap.items():
for ii in range(len(donutList)):
# Get the intra- and extra-focal donut images
# Check the sensor is the corner WFS or not.
# Only consider "intra" WFS.
# For LsstSimMapper, Intra: C0 -> A; Extra: C1 -> B
# For LsstCamMapper, Intra: C0 -> SW1; Extra: C1 ->SW0
# Look for the intra-focal image
if sensorName.endswith("SW1"):
intraDonut = donutList[ii]
# Get the extra-focal sensor name
extraFocalSensorName = sensorName.replace("SW1", "SW0")
# Get the donut list of extra-focal sensor
extraDonutList = donutMap[extraFocalSensorName]
if ii < len(extraDonutList):
extraDonut = extraDonutList[ii]
else:
continue
# Pass the extra-focal image
elif sensorName.endswith("SW0"):
continue
# Scientific sensor
else:
intraDonut = extraDonut = donutList[ii]
# Calculate the wavefront error
# Get the field X, Y position
intraFieldXY = intraDonut.getFieldPos()
extraFieldXY = extraDonut.getFieldPos()
# Get the defocal images
intraImg = intraDonut.getIntraImg()
extraImg = extraDonut.getExtraImg()
# Calculate the wavefront error
zer4UpNm = self._calcSglWfErr(
intraImg, extraImg, intraFieldXY, extraFieldXY
)
# Put the value to the donut image
intraDonut.setWfErr(zer4UpNm)
extraDonut.setWfErr(zer4UpNm)
# Intentionally to expose this return value to show the input,
# donutMap, has been modified.
return donutMap
def _calcSglWfErr(self, intraImg, extraImg, intraFieldXY, extraFieldXY):
"""Calculate the wavefront error in annular Zernike polynomials
(z4-z22) for single donut.
Parameters
----------
intraImg : numpy.ndarray
Intra-focal donut image.
extraImg : numpy.ndarray
Extra-focal donut image.
intraFieldXY : tuple
Field x, y in degree of intra-focal donut image.
extraFieldXY : tuple
Field x, y in degree of extra-focal donut image.
Returns
-------
numpy.ndarray
Coefficients of Zernike polynomials (z4 - z22) in nm.
"""
# Set the images
self.wfEsti.setImg(intraFieldXY, DefocalType.Intra, image=intraImg)
self.wfEsti.setImg(extraFieldXY, DefocalType.Extra, image=extraImg)
# Reset the wavefront estimator
self.wfEsti.reset()
# Calculate the wavefront error
zer4UpNm = self.wfEsti.calWfsErr()
return zer4UpNm
def calcAvgWfErrOnSglCcd(self, donutList):
"""Calculate the average of wavefront error on single CCD.
CCD: Charge-coupled device.
Parameters
----------
donutList : list
List of donut object (type: DonutImage).
Returns
-------
numpy.ndarray
Average of wavefront error in nm.
"""
# Calculate the weighting of donut image
wgtRatio = self._calcWeiRatio(donutList)
# Calculate the mean wavefront error
avgErr = 0
for ii in range(len(donutList)):
donut = donutList[ii]
zer4UpNmArr = donut.getWfErr()
# Assign the zer4UpNm
if len(zer4UpNmArr) == 0:
zer4UpNm = 0
else:
zer4UpNm = zer4UpNmArr
avgErr = avgErr + wgtRatio[ii] * zer4UpNm
return avgErr
def _calcWeiRatio(self, donutList):
"""Calculate the weighting ratio of donut in the list.
Parameters
----------
donutList : list
List of donut object (type: DonutImage).
Returns
-------
numpy.ndarray
Array of weighting ratio of donuts to do the average of wavefront
error.
"""
# Weighting of donut image. Use the simple average at this moment.
# Need to consider the S/N and other factors in the future.
# Check the available zk and give the ratio
wgtRatio = []
for donut in donutList:
if len(donut.getWfErr()) == 0:
wgtRatio.append(0)
else:
wgtRatio.append(1)
# Do the normalization
wgtRatioArr = np.array(wgtRatio)
normalizedwgtRatioArr = wgtRatioArr / np.sum(wgtRatioArr)
return normalizedwgtRatioArr
def genMasterDonut(self, donutMap, zcCol=np.zeros(22)):
"""Generate the master donut map.
Parameters
----------
donutMap : dict
Donut image map. The dictionary key is the sensor name. The
dictionary item is the donut image (type: DonutImage).
zcCol : numpy.ndarray, optional
Coefficients of wavefront (z1-z22) in nm. (the default is
np.zeros(22).)
Returns
-------
dict
Master donut image map. The dictionary key is the sensor name. The
dictionary item is the master donut image (type: DonutImage).
"""
masterDonutMap = dict()
for sensorName, donutList in donutMap.items():
# Get the master donut on single CCD
masterDonut = self._genMasterImgOnSglCcd(donutList, zcCol=zcCol)
# Put the master donut to donut map
masterDonutMap[sensorName] = [masterDonut]
return masterDonutMap
def _genMasterImgOnSglCcd(self, donutList, zcCol):
"""Generate the master donut image on single CCD.
CCD: Charge-coupled device.
Parameters
----------
sensorName : str
Canonical sensor name (e.g. "R:2,2 S:1,1").
donutList : list
List of donut object (type: DonutImage).
zcCol : numpy.ndarray
Coefficients of wavefront (z1-z22) in nm.
Returns
-------
DonutImage
Master donut.
"""
intraProjImgList = []
extraProjImgList = []
for donut in donutList:
# Get the field x, y
fieldXY = donut.getFieldPos()
# Set the image
intraImg = donut.getIntraImg()
if intraImg is not None:
# Get the projected image
projImg = self._getProjImg(fieldXY, DefocalType.Intra, intraImg, zcCol)
# Collect the projected donut
intraProjImgList.append(projImg)
extraImg = donut.getExtraImg()
if extraImg is not None:
# Get the projected image
projImg = self._getProjImg(fieldXY, DefocalType.Extra, extraImg, zcCol)
# Collect the projected donut
extraProjImgList.append(projImg)
# Generate the master donut
stackIntraImg = self._stackImg(intraProjImgList)
stackExtraImg = self._stackImg(extraProjImgList)
# Put the master donut to donut map
pixelX, pixelY = searchDonutPos(stackIntraImg)
masterDonut = DonutImage(
0, pixelX, pixelY, 0, 0, intraImg=stackIntraImg, extraImg=stackExtraImg
)
return masterDonut
def _getProjImg(self, fieldXY, defocalType, defocalImg, zcCol):
"""Get the projected image on the pupil.
Parameters
----------
fieldXY : tuple
Position of donut on the focal plane in the degree for intra- and
extra-focal images (field X, field Y).
defocalType : enum 'DefocalType'
Defocal type of image.
defocalImg : numpy.ndarray
Donut defocal image.
zcCol : numpy.ndarray
Coefficients of wavefront (z1-z22).
Returns
-------
numpy.ndarray
Projected image.
"""
# Set the image
self.wfEsti.setImg(fieldXY, defocalType, image=defocalImg)
# Get the image in the type of CompensationImageDecorator
if defocalType == DefocalType.Intra:
img = self.wfEsti.getIntraImg()
elif defocalType == DefocalType.Extra:
img = self.wfEsti.getExtraImg()
# Get the distortion correction (offaxis)
algo = self.wfEsti.getAlgo()
offAxisCorrOrder = algo.getOffAxisPolyOrder()
inst = self.wfEsti.getInst()
img.setOffAxisCorr(inst, offAxisCorrOrder)
# Do the image cocenter
img.imageCoCenter(inst)
# Do the compensation/ projection
img.compensate(inst, algo, zcCol, self.wfEsti.getOptModel())
# Return the projected image
return img.getImg()
def _stackImg(self, imgList):
"""Stack the images.
Parameters
----------
imgList : list
List of image.
Returns
-------
numpy.ndarray
Stacked image.
"""
if len(imgList) == 0:
stackImg = None
else:
# Get the minimun image dimension
dimXlist = []
dimYlist = []
for img in imgList:
dy, dx = img.shape
dimXlist.append(dx)
dimYlist.append(dy)
dimX = np.min(dimXlist)
dimY = np.min(dimYlist)
deltaX = dimX // 2
deltaY = dimY // 2
# Stack the image by summation directly
stackImg = np.zeros([dimY, dimX])
for img in imgList:
dy, dx = img.shape
cy = int(dy / 2)
cx = int(dx / 2)
stackImg += img[cy - deltaY : cy + deltaY, cx - deltaX : cx + deltaX]
return stackImg
class WepController(object):
CORNER_WFS_LIST = ["R:0,0 S:2,2,A", "R:0,0 S:2,2,B", "R:0,4 S:2,0,A",
"R:0,4 S:2,0,B", "R:4,0 S:0,2,A", "R:4,0 S:0,2,B",
"R:4,4 S:0,0,A", "R:4,4 S:0,0,B"]
def __init__(self, dataCollector, isrWrapper, sourSelc, sourProc, wfEsti):
"""Initialize the wavefront estimation pipeline (WEP) controller class.
Parameters
----------
dataCollector : CamDataCollector
Camera data collector.
isrWrapper : CamIsrWrapper
Instrument signature removal (ISR) wrapper.
sourSelc : SourceSelector
Source selector.
sourProc : SourceProcessor
Source processor.
wfEsti : WfEstimator
Wavefront estimator.
"""
self.dataCollector = dataCollector
self.isrWrapper = isrWrapper
self.sourSelc = sourSelc
self.sourProc = sourProc
self.wfEsti = wfEsti
self.butlerWrapper = None
def getDataCollector(self):
"""Get the attribute of data collector.
Returns
-------
CamDataCollector
Data collector.
"""
return self.dataCollector
def getIsrWrapper(self):
"""Get the attribute of ISR wraper.
ISR: Instrument signature removal.
Returns
-------
CamIsrWrapper
ISR wraper.
"""
return self.isrWrapper
def getSourSelc(self):
"""Get the attribute of source selector.
Returns
-------
SourceSelector
Source selector.
"""
return self.sourSelc
def getSourProc(self):
"""Get the attribute of source processor.
Returns
-------
SourceProcessor
Source processor.
"""
return self.sourProc
def getWfEsti(self):
"""Get the attribute of wavefront estimator.
Returns
-------
WfEstimator
Wavefront estimator.
"""
return self.wfEsti
def getButlerWrapper(self):
"""Get the attribute of butler wrapper.
Returns
-------
ButlerWrapper
Butler wrapper.
"""
return self.butlerWrapper
def setPostIsrCcdInputs(self, inputs):
"""Set inputs of post instrument signature removal (ISR) CCD images.
Parameters
----------
inputs : RepositoryArgs, dict, or str
Can be a single item or a list. Provides arguments to load an
existing repository (or repositories). String is assumed to be a
URI and is used as the cfgRoot (URI to the location of the cfg
file). (Local file system URI does not have to start with
'file://' and in this way can be a relative path). The
'RepositoryArgs' class can be used to provide more parameters with
which to initialize a repository (such as 'mapper', 'mapperArgs',
'tags', etc. See the 'RepositoryArgs' documentation for more
details). A dict may be used as shorthand for a 'RepositoryArgs'
class instance. The dict keys must match parameters to the
'RepositoryArgs.__init__' function.
"""
self.butlerWrapper = ButlerWrapper(inputs)
def getPostIsrImgMapByPistonDefocal(self, sensorNameList, obsIdList):
"""Get the post ISR image map that the defocal images are by the
pistion motion.
Parameters
----------
sensorNameList : list
List of sensor name.
obsIdList : list
Observation Id list in [intraObsId, extraObsId].
Returns
-------
dict
Post-ISR image map. The dictionary key is the sensor name. The
dictionary item is the defocal image on the camera coordinate.
(type: DefocalImage).
"""
# Get the waveront image map
wfsImgMap = dict()
for sensorName in sensorNameList:
# Get the sensor name information
raft, sensor = self._getSensorInfo(sensorName)[0:2]
# The intra/ extra defocal images are decided by obsId
imgList = []
for visit in obsIdList:
# Get the exposure image in ndarray
exp = self.butlerWrapper.getPostIsrCcd(int(visit), raft, sensor)
img = self.butlerWrapper.getImageData(exp)
# Transform the image in DM coordinate to camera coordinate.
camImg = self._transImgDmCoorToCamCoor(img)
# Collect the image
imgList.append(camImg)
wfsImgMap[sensorName] = DefocalImage(intraImg=imgList[0],
extraImg=imgList[1])
return wfsImgMap
def _transImgDmCoorToCamCoor(self, dmImg):
"""Transfrom the image in DM coordinate to camera coordinate.
Parameters
----------
dmImg : numpy.ndarray
Image in DM coordinate.
Returns
-------
numpy.ndarray
Image is camera coordinate.
"""
# The relationship between DM and camera coordinate should be 90 degree
# difference.
# For the PhoSim mapper, it is the transpose (x, y) to (y, x). This
# should be fixed.
camImg = dmImg.T
return camImg
def getPostIsrImgMapOnCornerWfs(self, sensorNameList, obsId):
"""Get the post ISR image map of corner wavefront sensors.
Parameters
----------
sensorNameList : list
List of sensor name.
obsId : int
Observation Id.
Returns
-------
dict
Post-ISR image map. The dictionary key is the sensor name. The
dictionary item is the defocal image on the camera coordinate.
(type: DefocalImage).
"""
pass
def _getSensorInfo(self, sensorName):
"""Get the sensor information.
Parameters
----------
sensorName : str
Sensor name (e.g. "R:2,2 S:1,1" or "R:0,0 S:2,2,A")
Returns
-------
str
Raft.
str
Sensor.
str
Channel.
"""
raft = sensor = channel = None
# Use the regular expression to analyze the input name
m = re.match(r"R:(\d,\d) S:(\d,\d)(?:,([A,B]))?$", sensorName)
if (m is not None):
raft, sensor, channel = m.groups()[0:3]
# This is for the phosim mapper use.
# For example, raft is "R22" and sensor is "S11".
raftAbbName = "R" + raft[0] + raft[-1]
sensorAbbName = "S" + sensor[0] + sensor[-1]
return raftAbbName, sensorAbbName, channel
def getDonutMap(self, neighborStarMap, wfsImgMap, filterType,
doDeblending=False):
"""Get the donut map on each wavefront sensor (WFS).
Parameters
----------
neighborStarMap : dict
Information of neighboring stars and candidate stars with the name
of sensor as a dictionary.
wfsImgMap : dict
Post-ISR image map. The dictionary key is the sensor name. The
dictionary item is the defocal image on the camera coordinate.
(type: DefocalImage).
filterType : FilterType
Filter type.
doDeblending : bool, optional
Do the deblending or not. If False, only consider the single donut
based on the bright star catalog.(the default is False.)
Returns
-------
dict
Donut image map. The dictionary key is the sensor name. The
dictionary item is the list of donut image (type:
list[DonutImage]).
"""
donutMap = dict()
for sensorName, nbrStar in neighborStarMap.items():
# Get the abbraviated sensor name
abbrevName = abbrevDectectorName(sensorName)
# Configure the source processor
self.sourProc.config(sensorName=abbrevName)
# Get the defocal images: [intra, extra]
defocalImgList = [wfsImgMap[sensorName].getIntraImg(),
wfsImgMap[sensorName].getExtraImg()]
# Get the bright star id list on specific sensor
brightStarIdList = list(nbrStar.getId())
for starIdIdx in range(len(brightStarIdList)):
# Get the single star map
for jj in range(len(defocalImgList)):
ccdImg = defocalImgList[jj]
# Get the segment of image
if (ccdImg is not None):
singleSciNeiImg, allStarPosX, allStarPosY, magRatio, \
offsetX, offsetY = \
self.sourProc.getSingleTargetImage(
ccdImg, nbrStar, starIdIdx, filterType)
# Only consider the single donut if no deblending
if (not doDeblending) and (len(magRatio) != 1):
continue
# Get the single donut/ deblended image
if (len(magRatio) == 1) or (not doDeblending):
imgDeblend = singleSciNeiImg
if (len(magRatio) == 1):
realcx, realcy = searchDonutPos(imgDeblend)
else:
realcx = allStarPosX[-1]
realcy = allStarPosY[-1]
# Do the deblending or not
elif (len(magRatio) == 2 and doDeblending):
imgDeblend, realcx, realcy = \
self.sourProc.doDeblending(
singleSciNeiImg, allStarPosX, allStarPosY,
magRatio)
# Update the magnitude ratio
magRatio = [1]
else:
continue
# Extract the image
if (len(magRatio) == 1):
sizeInPix = self.wfEsti.getSizeInPix()
x0 = np.floor(realcx - sizeInPix / 2).astype("int")
y0 = np.floor(realcy - sizeInPix / 2).astype("int")
imgDeblend = imgDeblend[y0:y0 + sizeInPix,
x0:x0 + sizeInPix]
# Rotate the image if the sensor is the corner
# wavefront sensor
if sensorName in self.CORNER_WFS_LIST:
# Get the Euler angle
eulerZangle = round(self.sourProc.getEulerZinDeg(
abbrevName))
# Change the sign if the angle < 0
while (eulerZangle < 0):
eulerZangle += 360
# Do the rotation of matrix
numOfRot90 = eulerZangle // 90
imgDeblend = np.flipud(
np.rot90(np.flipud(imgDeblend), numOfRot90))
# Put the deblended image into the donut map
if sensorName not in donutMap.keys():
donutMap[sensorName] = []
# Check the donut exists in the list or not
starId = brightStarIdList[starIdIdx]
donutIndex = self._searchDonutListId(
donutMap[sensorName], starId)
# Create the donut object and put into the list if it
# is needed
if (donutIndex < 0):
# Calculate the field X, Y
pixelX = realcx + offsetX
pixelY = realcy + offsetY
fieldX, fieldY = self.sourProc.camXYtoFieldXY(
pixelX, pixelY)
# Instantiate the DonutImage class
donutImg = DonutImage(starId, pixelX, pixelY,
fieldX, fieldY)
donutMap[sensorName].append(donutImg)
# Search for the donut index again
donutIndex = self._searchDonutListId(
donutMap[sensorName], starId)
# Get the donut image list
donutList = donutMap[sensorName]
# Take the absolute value for images, which might
# contain the negative value after the ISR correction.
# This happens for the amplifier images.
imgDeblend = np.abs(imgDeblend)
# Set the intra focal image
if (jj == 0):
donutList[donutIndex].setImg(intraImg=imgDeblend)
# Set the extra focal image
elif (jj == 1):
donutList[donutIndex].setImg(extraImg=imgDeblend)
return donutMap
def _searchDonutListId(self, donutList, starId):
"""Search the bright star ID in the donut list.
Parameters
----------
donutList : list
List of DonutImage object.
starId : int
Star Id.
Returns
-------
int
Index of donut image object with the specific starId.
"""
index = -1
for ii in range(len(donutList)):
if (donutList[ii].getStarId() == int(starId)):
index = ii
break
return index
def calcWfErr(self, donutMap):
"""Calculate the wavefront error in annular Zernike polynomials
(z4-z22).
Parameters
----------
donutMap : dict
Donut image map. The dictionary key is the sensor name. The
dictionary item is the donut image (type: DonutImage).
Returns
-------
dict
Donut image map with calculated wavefront error.
"""
for sensorName, donutList in donutMap.items():
for ii in range(len(donutList)):
# Get the intra- and extra-focal donut images
# Check the sensor is the corner WFS or not. Only consider "A"
# Intra: C0 -> A; Extra: C1 -> B
# Look for the intra-focal image
if sensorName.endswith("A"):
intraDonut = donutList[ii]
# Get the extra-focal sensor name
extraFocalSensorName = sensorName.replace("A", "B")
# Get the donut list of extra-focal sensor
extraDonutList = donutMap[extraFocalSensorName]
if (ii < len(extraDonutList)):
extraDonut = extraDonutList[ii]
else:
continue
# Pass the extra-focal image
elif sensorName.endswith("B"):
continue
# Scientific sensor
else:
intraDonut = extraDonut = donutList[ii]
# Calculate the wavefront error
# Get the field X, Y position
intraFieldXY = intraDonut.getFieldPos()
extraFieldXY = extraDonut.getFieldPos()
# Get the defocal images
intraImg = intraDonut.getIntraImg()
extraImg = extraDonut.getExtraImg()
# Calculate the wavefront error
zer4UpNm = self._calcSglWfErr(intraImg, extraImg, intraFieldXY,
extraFieldXY)
# Put the value to the donut image
intraDonut.setWfErr(zer4UpNm)
extraDonut.setWfErr(zer4UpNm)
# Intentionally to expose this return value to show the input,
# donutMap, has been modified.
return donutMap
def _calcSglWfErr(self, intraImg, extraImg, intraFieldXY, extraFieldXY):
"""Calculate the wavefront error in annular Zernike polynomials
(z4-z22) for single donut.
Parameters
----------
intraImg : numpy.ndarray
Intra-focal donut image.
extraImg : numpy.ndarray
Extra-focal donut image.
intraFieldXY : tuple
Field x, y in degree of intra-focal donut image.
extraFieldXY : tuple
Field x, y in degree of extra-focal donut image.
Returns
-------
numpy.ndarray
Coefficients of Zernike polynomials (z4 - z22) in nm.
"""
# Set the images
self.wfEsti.setImg(intraFieldXY, DefocalType.Intra, image=intraImg)
self.wfEsti.setImg(extraFieldXY, DefocalType.Extra, image=extraImg)
# Reset the wavefront estimator
self.wfEsti.reset()
# Calculate the wavefront error
zer4UpNm = self.wfEsti.calWfsErr()
return zer4UpNm
def calcAvgWfErrOnSglCcd(self, donutList):
"""Calculate the average of wavefront error on single CCD.
CCD: Charge-coupled device.
Parameters
----------
donutList : list
List of donut object (type: DonutImage).
Returns
-------
numpy.ndarray
Average of wavefront error in nm.
"""
# Calculate the weighting of donut image
wgtRatio = self._calcWeiRatio(donutList)
# Calculate the mean wavefront error
avgErr = 0
for ii in range(len(donutList)):
donut = donutList[ii]
zer4UpNmArr = donut.getWfErr()
# Assign the zer4UpNm
if (len(zer4UpNmArr) == 0):
zer4UpNm = 0
else:
zer4UpNm = zer4UpNmArr
avgErr = avgErr + wgtRatio[ii] * zer4UpNm
return avgErr
def _calcWeiRatio(self, donutList):
"""Calculate the weighting ratio of donut in the list.
Parameters
----------
donutList : list
List of donut object (type: DonutImage).
Returns
-------
numpy.ndarray
Array of weighting ratio of donuts to do the average of wavefront
error.
"""
# Weighting of donut image. Use the simple average at this moment.
# Need to consider the S/N and other factors in the future.
# Check the available zk and give the ratio
wgtRatio = []
for donut in donutList:
if (len(donut.getWfErr()) == 0):
wgtRatio.append(0)
else:
wgtRatio.append(1)
# Do the normalization
wgtRatioArr = np.array(wgtRatio)
normalizedwgtRatioArr = wgtRatioArr / np.sum(wgtRatioArr)
return normalizedwgtRatioArr
def genMasterDonut(self, donutMap, zcCol=np.zeros(22)):
"""Generate the master donut map.
Parameters
----------
donutMap : dict
Donut image map. The dictionary key is the sensor name. The
dictionary item is the donut image (type: DonutImage).
zcCol : numpy.ndarray, optional
Coefficients of wavefront (z1-z22) in nm. (the default is
np.zeros(22).)
Returns
-------
dict
Master donut image map. The dictionary key is the sensor name. The
dictionary item is the master donut image (type: DonutImage).
"""
masterDonutMap = dict()
for sensorName, donutList in donutMap.items():
# Get the master donut on single CCD
masterDonut = self._genMasterImgOnSglCcd(donutList, zcCol=zcCol)
# Put the master donut to donut map
masterDonutMap[sensorName] = [masterDonut]
return masterDonutMap
def _genMasterImgOnSglCcd(self, donutList, zcCol):
"""Generate the master donut image on single CCD.
CCD: Charge-coupled device.
Parameters
----------
sensorName : str
Canonical sensor name (e.g. "R:2,2 S:1,1").
donutList : list
List of donut object (type: DonutImage).
zcCol : numpy.ndarray
Coefficients of wavefront (z1-z22) in nm.
Returns
-------
DonutImage
Master donut.
"""
intraProjImgList = []
extraProjImgList = []
for donut in donutList:
# Get the field x, y
fieldXY = donut.getFieldPos()
# Set the image
intraImg = donut.getIntraImg()
if (intraImg is not None):
# Get the projected image
projImg = self._getProjImg(fieldXY, DefocalType.Intra,
intraImg, zcCol)
# Collect the projected donut
intraProjImgList.append(projImg)
extraImg = donut.getExtraImg()
if (extraImg is not None):
# Get the projected image
projImg = self._getProjImg(fieldXY, DefocalType.Extra,
extraImg, zcCol)
# Collect the projected donut
extraProjImgList.append(projImg)
# Generate the master donut
stackIntraImg = self._stackImg(intraProjImgList)
stackExtraImg = self._stackImg(extraProjImgList)
# Put the master donut to donut map
pixelX, pixelY = searchDonutPos(stackIntraImg)
masterDonut = DonutImage(0, pixelX, pixelY, 0, 0,
intraImg=stackIntraImg, extraImg=stackExtraImg)
return masterDonut
def _getProjImg(self, fieldXY, defocalType, defocalImg, zcCol):
"""Get the projected image on the pupil.
Parameters
----------
fieldXY : tuple
Position of donut on the focal plane in the degree for intra- and
extra-focal images (field X, field Y).
defocalType : enum 'DefocalType'
Defocal type of image.
defocalImg : numpy.ndarray
Donut defocal image.
zcCol : numpy.ndarray
Coefficients of wavefront (z1-z22).
Returns
-------
numpy.ndarray
Projected image.
"""
# Set the image
self.wfEsti.setImg(fieldXY, defocalType, image=defocalImg)
# Get the image in the type of CompensationImageDecorator
if (defocalType == DefocalType.Intra):
img = self.wfEsti.getIntraImg()
elif (defocalType == DefocalType.Extra):
img = self.wfEsti.getExtraImg()
# Get the distortion correction (offaxis)
algo = self.wfEsti.getAlgo()
offAxisCorrOrder = algo.getOffAxisPolyOrder()
inst = self.wfEsti.getInst()
img.setOffAxisCorr(inst, offAxisCorrOrder)
# Do the image cocenter
img.imageCoCenter(inst)
# Do the compensation/ projection
img.compensate(inst, algo, zcCol, self.wfEsti.getOptModel())
# Return the projected image
return img.getImg()
def _stackImg(self, imgList):
"""Stack the images.
Parameters
----------
imgList : list
List of image.
Returns
-------
numpy.ndarray
Stacked image.
"""
if (len(imgList) == 0):
stackImg = None
else:
# Get the minimun image dimension
dimXlist = []
dimYlist = []
for img in imgList:
dy, dx = img.shape
dimXlist.append(dx)
dimYlist.append(dy)
dimX = np.min(dimXlist)
dimY = np.min(dimYlist)
deltaX = dimX//2
deltaY = dimY//2
# Stack the image by summation directly
stackImg = np.zeros([dimY, dimX])
for img in imgList:
dy, dx = img.shape
cy = int(dy / 2)
cx = int(dx / 2)
stackImg += img[cy - deltaY:cy + deltaY,
cx - deltaX:cx + deltaX]
return stackImg
def poltExposureImage(exposure,
name="",
scale="log",
cmap="gray",
vmin=None,
vmax=None,
saveFilePath=None):
"""Plot the exposure image.
Parameters
----------
exposure : Exposure
Data butler of exposure image.
name : str, optional
Image title name. (the default is "".)
scale : str, optional
Scale of image map (log or linear). (the default is "log".)
cmap : str, optional
Color map definition. (the default is "gray".)
vmin : float, optional
Mininum value to show. This normalizes the luminance data. (the default
is None.)
vmax : float, optional
Maximum value to show. This normalizes the luminance data. (the default
is None.)
saveFilePath : str, optional
Save image to file path. (the default is None.)
"""
# Get the image data
img = ButlerWrapper.getImageData(exposure)
# Change the scale if needed
if scale not in ("linear", "log"):
print(
"No %s scale to choose. Only 'linear' and 'log' scales are allowed."
% scale)
return
# Decide the norm in imshow for the ploting
if scale == "linear":
plotNorm = None
elif scale == "log":
if (img.min()) < 0:
plotNorm = SymLogNorm(linthresh=0.03)
else:
plotNorm = LogNorm()
# Plot the image
plt.figure()
plt.imshow(img,
cmap=cmap,
origin="lower",
norm=plotNorm,
vmin=vmin,
vmax=vmax)
plt.colorbar()
plt.title(name)
if saveFilePath is None:
plt.show()
else:
plt.savefig(saveFilePath, bbox_inches="tight")
plt.close()