In this page processors available for image analysis are described. All processors required for standard image reduction are already available. Other, more advanced tools will be added in the future.
- luiza::FitsImageReader
This processor can be used to read images from fit files. List of file names or name of the file containing file names should be specified. By default one image is read for each analysis pass, but this can be changed by setting ImagesPerLoop parameter. Read images are stored in an image collection with name given by ImageCollectionName parameter.
- luiza::SimpleImageAdder
This processor can be used to add images stored in input collection and store resulting frames into output collection. User has to specify InputCollectionName and OutputCollectionName. The default behaviour is to add all images in input collection (in given analysis loop) into one output image. This can be changed using NumberAddImages and UseManyLoops parameters. User can also decide to average images instead of adding them, by specifying AverageImages flag.
- luiza::ImageMedian
Processor to calculate median image from images in the input collection. It is an alternative to simple adding the images, eg. in case of dark frames. All images from the given collection are used to calculate median image.
- luiza::ImageNormalizationProcessor
This processor can be used to normalize raw images ie. subtract dark and bias images, and divide by flat frame. Corrected frames are stored in a new collection, if specified, otherwise old images are modified. Normalizaction of flat frames (if raw images are used) is also possible, as is the automatic weigth calculation for dark and bias images, based on raw image and dar image exposure times (see processor description for details).
- luiza::SimpleImageRotator
Simple processor implementing basic image geometry manipulations. Includes image rotation (by 90, 180 or 270 degree), image flipping (reflection with respect to horizontal or vertical axis), as well as cropping and rebinning.
- luiza::PixelClusterFinder
Simple processor to find pixel clusters (ie. stars) on the CCD frame. This algorithm searches for groups of pixels with signal above the defined threshold. The seed pixel is searched first (with tighter seed selection cut) and then the neighbouring pixels, with signal above pixel signal threshold (which can be looser) are added to form the cluster. The algroithm assumes that the cluster should have one maximum only, if two maxima (above seed threshold) are found, two clusters will be reconstructed. To define thresholds, background level and average noise level are calculated first. For images with significant background variation over the frame, background level can be calculated for smaller subframes and then interpolated (linear and 2nd order interpolation available).
- luiza::ImageLaplace
Processor to apply Laplace transformation to images. Laplace transformation can be used to "sharpen" the image and remove (variable) background contribution. This can improve object finding efficiency in some cases. However, precision of object brightness determination on the resulting image is largly reduced. After finding objects on the transformed image one should consider using aperture photometry on the oryginal image (see examples).
- luiza::FastAperturePhotometry
Processor to calculate aperture photometry on CCD images. If object list is specified on input, photometry is calculated for all objects in the list and results added to the table. For the targeted analysis, coordinates of the objects of interest can be specified. However, valid astrometry has to be found in the image fits header in that case.
- luiza::WeightedAperturePhotometry
Processor to calculate aperture photometry on CCD images in the input collection. If input object collection is specified, photometry is calculated for all objects in the object list. For the targeted analysis, coordinates of the objects of interest can be specified. However, valid astrometry has to be found in the image fits header in that case. Two approaches for background calculation are possible: either it is calculated from media of pixels in annulus around object position (when AnnulusRadii is specified) or from the background map calculated for the whole image (as in luiza::BackgroundImage processor). The algorithm used differs from the one implemented in luiza::FastAperturePhotometry by introducing weights for pixels which are at the aperture boundary. The weight depends also on the exact position of the object.
- luiza::BackgroundImage
Processor calculates background map from given image. The frame is divided into FrameDivision x FrameDivision subframes, then background level and noise are calculated from median in each subframe. Finally, background and/or noise image is created by interpolating between subframe centers. New images have the same resolution as the input image.
- luiza::RootImageViewer
Processor allowing to view images during processing. This processor uses CERN root classes to display images from collection given by ImageCollectionName parameter. Images can be displayed as images or as 2D histograms. In the later case it is possible to overlay objects from the give object list (eg. to check performance of object findind algorithm)
- luiza::FitsImageWriter
This processor can be used to write modified images or new images resulting from the analysis to the standard fit files. User should specify the file name root (FitsFileNameRoot parameter) - it will be appended by image number and ".fit" extension. All images read from an input image collection given by ImageCollectionName are stored to file. Format of the output file can be set with BitsPerPixel parameter.
- luiza::TestProcessor
This processor prints out all collections and dimensions of all images found. Can be used to monitor performance of other processors.
- luiza::ExampleProcessor
This processor prints out image statistics from given collection. It can be compiled as a separate library loaded dynamicly by Luiza. It is to be used as a template for developing user processors.
1.8.9.1
