Image Integration

After an image session we have a big set of data files consisting of bias-, dark-, flat- and light images. All the light images have to be calibrated with the bias-, dark-, flat images. After calibration they need to be registered and stacked in order to get our final (unprocessed) image. Before we can start with the callibration process we need to create the masterbias-, masterdark- and masterflat files.

Creating master files

The master files can easily be made with the "ImageIntegration" process in PixInsight according to the following table.

In PixInsight this will look like :

Hit the "Apply Global" button in the low left corner (small circle) to apply the integration process for each dataset and save the resulting master files.

Calibrating en registering the light frames

This process can be done with the "Batch Preprocessing Script". Load the masterbias, masterdark and masterflat in the corresponding tabs of the script and select the "Use master bias", "Use master dark" and "Use master flat" boxes on the right.
Then load the light frames and select the "Registration Reference Image" and "Output Directory", the check box for "CFA images" should only be selected when using OSC images (leave it unchecked for monochrome images). Check the "Generate drizzle data" checkbox and uncheck "Apply" checkbox under the "Image Integration" tab.

Hit the "Run" button, all light frames will be calibrated, registered and saved under the selected "Output directory" (including the drizzle data).

Image Integrating

Depending on the number of light frames taken a few integrating techniques can be used for an optimal result, see this presentation for more details. Following this presentation the next table gives an overview for the different techniques.

Open now the "ImageIntegration" process and first add the registered light frames, after that add also the drizzle files.

After all files are loaded we can define the different fields in the "ImageIntegration" process. First of all we need to define the way we want to combine our images. As shown in the table above we can leave it to the following values in all cases.

The difference will be in the pixel rejection part of the process.

Integration of an small number of images (n<6) Integration of an mid number of images (6<n<10)
Integration of an mid number of images (10<n<15) Integration of an large number of images (n>15)

For each of the cases you can now experiment with the "low" and "high" limits in order to achieve the best ∆SNR (can be checked in the process control screen in PixInsight as described in the presentation).

After pressing the "Apply Global" button on the lower left (small blue circle) the final image will be calculated. The result is a, still, unprocessed final image stack. First save the image as a PixInsight *.xisf file. The result image is now ready for processing.

© Copyright Rob Kantelberg
free hit counters