IPP Software Navigation Tools IPP Links Communication Pan-STARRS Links

Changeset 4900


Ignore:
Timestamp:
Aug 29, 2005, 4:37:55 PM (21 years ago)
Author:
eugene
Message:

adding pstasks, updating psphot.tex

Location:
trunk/doc
Files:
11 added
3 edited

Legend:

Unmodified
Added
Removed

  • trunk/doc/Makefile

    r2501 r4900  
    99        @rm -f *~ #*
    1010        for i in $(DIR); do (cd $$i; make clean); done
     11
     12psdc:
     13        rsync -e ssh -auv PSDC-4xx/ poiserver0:doc/panstarrs/www/src/project/PSDC/PSDC-4xx/

  • trunk/doc/design/psched-tasks.txt

    r4895 r4900  
    1 
    2 
     1\documentclass[panstarrs,spec]{panstarrs}
     2
     3\title{PStask \& Metadata DB interactions} % put in your title
     4\subtitle{Job Flow in IPP}
     5\author{Eugene Magnier}
     6\audience{IPP}
     7%\shorttitle{PSPhot}
     8\group{Pan-STARRS IPP}
     9\project{Pan-STARRS IPP}
     10\organization{Institute for Astronomy}
     11\version{DR}
     12\docnumber{PSDC-xxx-xxx}
     13
     14\newcommand\ugriz{$u^\prime g^\prime r^\prime i^\prime z^\prime$}
     15\newcommand\grizy{$g r i z y$}
     16
     17\begin{document}
     18\maketitle
    319
    420This collection of diagrams shows the IPP tasks and the MDDB tables

  • trunk/doc/psphot/psphot.tex

    r4896 r4900  
    5151data rate.  The prototype telescope alone is expected to produce
    5252typically $\sim 700$ GB per night of imaging data.  These images will
    53 not be limited to high galactic lattitudes, so large numbers of
     53not be limited to high galactic latitudes, so large numbers of
    5454measurable stars can be expected in much of the data.  The combination
    5555of the high precision goals of the astrometric and photometric
    … …  
    8383\item Sextractor : pure aperture measurement with rudimentary
    8484  object subtraction.  pro: fast, widely used, easy to automate.  con:
    85   poor object separation in crowded regions, psf-modelling is only
     85  poor object separation in crowded regions, PSF-modeling is only
    8686  beta (psfex), what models are available?
    8787
    … …  
    8989  con: IRAF-based, aperture photometry.
    9090
    91 \item galfit : detailed galaxy modelling.  not a multi-object PSF
     91\item galfit : detailed galaxy modeling.  not a multi-object PSF
    9292  analysis tool.  con: does not provide a PSF model, not easily
    9393  automated.  very detailed results in very slow processing.  only a
    … …  
    112112Perl (or potentially Python). 
    113113
    114 \note{Discussion of the lessons learned from experience with previous
    115   analysis programs.  1) Flexible PSF model: functional form should be
    116   easily modified.  2) PSF variation is fundamental : PSF
    117   representation should incorporate 2-D variations.  3) Speed fitting
    118   with accurate parameter guesses.  3) Make good use of moment
    119   information to speed analysis.  4) careful definition of PSF
    120   validity tests.  5) careful analysis of aperture corrections.  6)
    121   flexible non-PSF models.  7) Good code abstraction to simplify
    122   modification. }
     114\note{Add discussion of the lessons learned from experience with previous
     115  analysis programs}
     116
     117\begin{itemize}
     118\item Flexible PSF model: functional form should be
     119  easily modified. 
     120\item PSF variation is fundamental : PSF representation should incorporate 2-D variations. 
     121\item Speed fitting with accurate parameter guesses. 
     122\item Make good use of moment information to speed analysis. 
     123\item Careful definition of PSF validity tests. 
     124\item Careful analysis of aperture corrections. 
     125\item Flexible non-PSF models. 
     126\item Good code abstraction to simplify modification.
     127\end{itemize}
    123128
    124129\section{Description of the PSPhot analysis steps}
    … …  
    189194
    190195The noise image, if not supplied is constructed by default from the
    191 flux image using the configuration supplied values of GAIN and
    192 READ_NOISE to calculate the appropriate Poisson statistics for each
    193 pixel.  In this case, the image is assumed to represent the readout
    194 from a single detector, with well-defined gain and read noise
     196flux image using the configuration supplied values of \code{GAIN} and
     197\code{READ_NOISE} to calculate the appropriate Poisson statistics for
     198each pixel.  In this case, the image is assumed to represent the
     199readout from a single detector, with well-defined gain and read noise
    195200characteristics.  In some obvious cases, this assumption will not be
    196201valid.  For example, if the input flux image is the result of an image
    197202stack with significantly variable number of input measurements per
    198203pixel, it will necessary to supply a noise image which accurately
    199 represents the noise as a function of position in the image. 
    200 
    201 \subsubsection{Intial Object Detection}
     204represents the noise as a function of position in the image.
     205
     206\subsubsection{Initial Object Detection}
    202207
    203208The objects are initially detected by finding the location of local
    … …  
    231236Once a collection of peaks have been identified, basic properties of
    232237the objects are measured.  First, the local sky flux is measured
    233 (using Median? user-specifie method?) within a square annulus with
     238(using Median? user-specific method?) within a square annulus with
    234239user-defined dimensions (\code{INNER_RADIUS} and \code{OUTER_RADIUS}).
    235240\note{rejection of some peaks based on the local sky measurement?}.
    … …  
    250255PSPhot uses an analytical model to represent the shape and flux of an
    251256object.  An important concept within the PSPhot code is the
    252 distinction bewteen a model which describes an object on an image and
     257distinction between a model which describes an object on an image and
    253258a model with describes the point-spread-function across an image. 
    254259
    … …  
    258263($x_o, y_o$), the elliptical shape parameters ($\sigma_x, \sigma_y,
    259264\sigma_{xy}$), the model normalization ($I_o$) and the local value of
    260 the background ($S$).  A specific object will have a partiular set of
     265the background ($S$).  A specific object will have a particular set of
    261266values for these different parameters.
    262267
    … …  
    278283each a function of the object centroid coordinates:
    279284\begin{eqnarray}
    280 \sigma_x    & = & f1(x,y) \\
    281 \sigma_y    & = & f2(x,y) \\
    282 \sigma_{xy} & = & f3(x,y) \\
     285\sigma_x    & = & f_1(x,y) \\
     286\sigma_y    & = & f_2(x,y) \\
     287\sigma_{xy} & = & f_3(x,y) \\
    283288\end{eqnarray}
    284289
    … …  
    314319these types of circumstances are abstracted, and a method is provided
    315320to return the necessary function to the higher-level software.  For
    316 example, each model type has its own function to define an intial
     321example, each model type has its own function to define an initial
    317322guess for the model, or a function to determine the radius for a given
    318323flux level.  These are then registered as part of the model function
    319324code.  Another function is then used to return the appropriate
    320325function for a specific model type.  For example, the
    321 psModelLookup_GetFunction will return the psModelLookup function for a
    322 given model type.  This mechanism makes it very easy to add new model
    323 functions into the PSPhot code base.  To add a new model function, the
    324 programmer simply defines a new model name (a string), the set of all
    325 necessary model lookup functions, and places the reference to the
    326 model code at the appropriate location in the psModelInit.c routine.
    327 It is not necessary to specify the PSF model functions independently
    328 or the object model functions.  Nor is it necessary to identify the
    329 intended use of a given object model function (ie, PSF-like object,
    330 galaxy, comet, etc).  Any model can be used for the PSF model.  The
    331 code currently uses a fixed translation between the object model
    332 parameters and the PSF model parameters.  It also defines a specific
    333 order for the 4 independent parameters.  \note{it may also require
    334 that two of the PSF-like parameters represent the shape in some way}.
     326\code{psModelLookup_GetFunction} will return the \code{psModelLookup}
     327function for a given model type.  This mechanism makes it very easy to
     328add new model functions into the PSPhot code base.  To add a new model
     329function, the programmer simply defines a new model name (a string),
     330the set of all necessary model lookup functions, and places the
     331reference to the model code at the appropriate location in the
     332psModelInit.c routine.  It is not necessary to specify the PSF model
     333functions independently or the object model functions.  Nor is it
     334necessary to identify the intended use of a given object model
     335function (ie, PSF-like object, galaxy, comet, etc).  Any model can be
     336used for the PSF model.  The code currently uses a fixed translation
     337between the object model parameters and the PSF model parameters.  It
     338also defines a specific order for the 4 independent parameters.
     339\note{it may also require that two of the PSF-like parameters
     340represent the shape in some way}.
    335341
    336342\subsubsection{PSF Object Candidate Selection}
    … …  
    339345identify a collection of objects in the image which are {\em likely}
    340346to be PSF-like.  PSPhot uses the object moments to make the initial
    341 guess at a colletion of PSF-like objects.  At this point, the program
    342 has measured the second order moments for all objects identified thier
     347guess at a collection of PSF-like objects.  At this point, the program
     348has measured the second order moments for all objects identified their
    343349peaks, as well as an approximate signal-to-noise ratio.  All objects
    344350with a S/N ratio greater than a user-defined parameter are selected by
    … …  
    353359the bin size is approximately 0.1 image pixels.  The binned $\sigma_x,
    354360\sigma_y$ plane is then examined to find a peak which has a
    355 significance greater than XXX.  Unless the image is extremely sparce,
     361significance greater than XXX.  Unless the image is extremely sparse,
    356362such a peak will be well-defined and should represent the objects
    357363which are all very similar in shape.  Other objects in the image will
    358364tend to land in very different locations, failing to produce a single
    359 peak.  To avoid detecting a peak from the unresoled cosmic rays,
     365peak.  To avoid detecting a peak from the unresolved cosmic rays,
    360366objects which have second-moments very close to 0 are ignored.  The
    361367only danger is if the PSF is very small and too many of these objects
    … …  
    381387iterative, and rejects the $3-\sigma$ outliers in each of three
    382388passes.  This fitting technique results in a robust measurement of the
    383 variation of the PSF model parametesr as a function of position
     389variation of the PSF model parameters as a function of position
    384390without being excessively biased by individual objects which fail
    385391drastically.  Objects whose model parameters are rejected by this
    … …  
    400406the PSF objects.  The difference between the aperture and fit
    401407magnitudes ({\em ApResid}) is a critical parameter for any PSF
    402 modelling software which uses an analytical model to represent the
     408modeling software which uses an analytical model to represent the
    403409flux distribution of the objects in an image. 
    404410
    … …  
    408414between two images.  Whether the goal is calibration of a science
    409415image taken at one location to a standard star image at another
    410 location, or the goal is simply the repetative photometry of the same
     416location, or the goal is simply the repetitive photometry of the same
    411417star at the same location in the image, it is always necessary to
    412418compare the photometry between two images.  If this measurement is to
    … …  
    450456Consider a typical bright object with a flux of (say) 40,000 counts in
    451457an image of background 1000 counts per pixel, with FWHM of 4 pixels.
    452 In principle, the flux of this object should be measureable with an
     458In principle, the flux of this object should be measurable with an
    453459accuracy of roughly 0.57\% ($\frac{\sqrt{40000 + 1000 \times
    45446012}}{40000}$).  However, the measurement of the sky is limited at some
    … …  
    572578
    573579PSPhot will use the user-selected galaxy model to attempt the galaxy
    574 model fits.  In the configuration system, the KEYWORD GAL_MODEL is set
    575 to the model of interest.  All suspected extended objects are fitted
    576 with the model, allowing all of the parameters to float.  The initial
    577 parameter guesses are critical here to achieving convergence on the
    578 model fits in a reasonable time.  The moments and the pixel flux
    579 distribution are used to make the initial parameter guess.  Many of
    580 the object parameters can be accurately guessed from the first and
     580model fits.  In the configuration system, the keyword \code{GAL_MODEL}
     581is set to the model of interest.  All suspected extended objects are
     582fitted with the model, allowing all of the parameters to float.  The
     583initial parameter guesses are critical here to achieving convergence
     584on the model fits in a reasonable time.  The moments and the pixel
     585flux distribution are used to make the initial parameter guess.  Many
     586of the object parameters can be accurately guessed from the first and
    581587second moments.  The power-law slope can be guessed by measuring the
    582588isophotal level at two elliptical radii and comparing the ratio to
    583 that expected. 
     589that expected.
    584590
    585591For each of the galaxy models (in fact for all object models), a
    … …  
    642648Finally, PSPhot can simply ignore the fitting process and instead
    643649glean information about the fainter sources on the basis of the peak
    644 charateristics.  In this option, the image is smoothed with the PSF
     650characteristics.  In this option, the image is smoothed with the PSF
    645651model, and the peak for each object is measured.  The peak flux and
    646652the local peak curvature theoretically give sufficient information to
    … …  
    675681will automatically result in inconsistent interpretation of the noise.
    676682
    677 For a difference image, both positive and negetive objects will be
     683For a difference image, both positive and negative objects will be
    678684present.  The basic peak detection algorithm will only trigger for the
    679685positive sources.  One solution is to simply apply PSPhot to both the
    … …  
    703709a 30 second exposure.  Even a main belt asteroid at roughly 1 AU would
    704710have reflect motion of approximately 1 degree per day, equivalent to
    705 1.25 arcsec in a 30 second exposure, and could be noticably smeared
     7111.25 arcsec in a 30 second exposure, and could be noticeably smeared
    706712and non-PSF-like.  A trailed-star model can be used to characterize
    707713these types of objects.  2) Small offset stars, either due to
Note: See TracChangeset for help on using the changeset viewer.