Jump to: navigation, search

Features in brief

  • Providing basic information about the images or other FITS data, pixel statistics.
  • Conversion of FITS images to more easily presentable formats.
  • Querying and modification of FITS header keywords.
  • Calibration of raw images, including the masking of bad, saturated or otherwise unuseable pixels.
  • Arithmetic operations on images, both on per-image and per-pixel basis.
  • Combination of multiple images into a single one.
  • Generic spatial geometric transformations of images (shifting, dilating, shrinking, clipping, higher order polynomial transformations, ...), including the registration of images to the same reference frame.
  • Generation of artificial images.
  • Detection and characterization of stellar profiles.
  • Coordinate list manipulation (fit and evaluation of geometric transformations, point matching, pair matching, cross-matching of catalogues and source lists, matching by identifiers, ...)
  • Instrumental photometry (aperture, image subtraction, analytic profile modelling, point-spread functions and various other sophisticated combinations of these).
  • Regression analysis and general numeric data manipulation.
  • The tasks are optimized for shell-level parallel processing.

Features by tasks


  • Basic arithmetic operations on images.
  • Operations on individual pixels.
  • Creation of images where the pixel intensities are are functions of the coordinates.


  • Performing simple arithmetic operations on a bunch of images.
  • Performing pixel mask modification on a bunch of images.
  • Trimming of a set of images.


  • Combination of a set of images into one image using various averaging and stacking methods.


  • Fitting optimal convolution transformation between a reference image and an input image.
  • Application of existing convolution transformation on a reference image.
  • Performing image subtraction using a reference image and an input image either by knowing the convolution transformation in advance or by fitting the optimal one during the process.
  • Management of so-called kernel files describing these convolution transformations.


  • Extracting various FITS keywords and the associated values from individual or a bunch of files.
  • Modification of FITS keywords in an individual or in a bunch of files.
  • Adding or removing FITS keywords to/from an individual or to/from a bunch of files.


  • Performing low-level manipulation of masks associated to FITS images.
  • Adding or removing masks (and various mask layers) to/from a FITS image.
  • Conversions between various mask types.
  • Adding or removing mask bits as the function of the pixel values or based on the characteristics of pixels.


  • Reporting a brief or a detailed summary about a FITS file, including primary image and various extensions (such as additional images, binary tables or textual tables).
  • Extracting raw information from FITS files, including dumping of image pixel data, mask bits, binary table data and textual table data.
  • Conversion of FITS images to PNM (PPM, PGM) format using
    • various scaling functions, such as linear, logarithmic, squared, square root, etc.,
    • scale limit determinations, such as minimum, maximum values, zscale, histrogram normalization, values derived from cumulative pixel intensity distributions, etc., and
    • various color palettes, including grey-scale, color, custom defined color as well as 8-bit and 16-bit dynamics.
  • Extracting simple pixel statistics from FITS image files.


  • Performing aperture photometry on normal images.
  • Performing aperture photometry on subtracted images by considering the information related to image convolution (if it has been applied in advance).
  • Determination of background level using the classic ways (considering pixels in annuli around the targets) as well as treating the background to be known in advance.
  • Support for arbitrary-shaped apertures using polygonal approximations by involving arbitrary number of boundary segments.
  • Conversion of fluxes to magnitudes.
  • Computation of flux and/or magnitude uncertainties by involving as many noise sources as possible (mainly background noise and photon noise).
  • Computation shape centroid coordinates and shape parameters, refinement of centroid coordinates.
  • Computation of the respective uncertainties of centroid coordinates and shape parameters by involving as many noise sources as possible.
  • Fitting and adjusting of large-scale flux variations on subtracted images considering a dependence on the coordinates and intrinsic colors in a form of arbitrary-order polynomials.
  • Treatment of nearby sources by removing them from the background determination.
  • Flagging the sources using the constraints defined by the masks associated to the FITS images.


  • Creation of images.
  • Simulation of various noise sources, such as background noise and photon noise.
  • Implantation of sources having various fluxes and shape parameters (including analytical shape models as well as external point-spread functions).
  • Quantization of pixel flux values.
  • Creation of random lists of objects.


  • Extracting list of point sources and the corresponding geometric parameters (centroid coordinates, shape paremeters, etc.).
  • Fitting of (optionally space-varying) point-spread function based on the detected sources, including sub-pixel level sampling.


  • Transformation of FITS images and the corresponding pixel masks.
  • Extractions of a single or multiple layers from a FITS data cube.


  • Transposing large amount of data based on unique keys found in the input data series.


  • Matching two data files using various criteria:
    • identifier matching (unique identifiers or keys are matched),
    • coordinate matching (points are matched based on Cartesian distance and topological correspondence)
    • point matching and the derivation of the corresponding analytic transformation (only in planar case)
  • Fine-tuning of point matching algorithms using various criteria


  • Derivation of best-fit spatial transformation between a matched list of coordinates.
  • Application of transformation data to a list of coordinates.
  • Application of astronomy-specific transformations on a list of coordinates.


  • Regression analysis and best-fit parameters estimation of input data by involving various algorithms, such as
    • linear least squares fitting;
    • non-linear least squares fitting, based on
      • Levenberg-Marquard procedure,
      • downhill simplex minimization,
      • Markov Chain Monte Carlo, etc.
    • Mapping of χ2 space
  • Uncertainty estimations by involving various procedures, such as
    • Linear error propagation and covariance matrix analysis,
    • Markov Chain Monte Carlo,
    • Bootsrapping, etc.
  • Evaluation of functions.
  • Fully symbolic operations, incuding analytical computation of partial derivatives.
  • Definition of additional functions during run-time.
  • User-defined and time-consuming functions can be implemented using shared libraries.
  • Various built-in functions useful in astronomical data analysis.
Personal tools