Understanding Calibration Groups

MaxIm DL uses a novel method for setting up the calibration frames, in that it allows the definition of multiple "groups" of files. This technique provides a high level of automation and flexibility.

As discussed above, calibration requires bias, dark, and flat files. In practice users typically take exposures with different settings in a single session. Individual exposures may vary by exposure time, CCD temperature, binning, filter band, subframe, and different cameras may even be in use (e.g. autoguider versus main camera).

MaxIm DL can perform exposure scaling to account for exposure duration (Auto Scale) and to some extent temperature (Auto Optimize). The software can also automatically extract a subframe from the calibration frames. But differences in binning, exposure bands, filters, and of course cameras cannot be accommodated except by using multiple calibration frames. Switching back and forth manually can be time consuming, particularly if complex exposure sequences are taken with different filters and binning, as is done for LRGB imaging and photometry.

A calibration "group" is simply a set of one or more calibration frames that are taken under the same conditions. A set of 10 dark frames taken at -20C with a specific camera would be one group. A set of flat frames taken with a green filter would be another.

These groups can be automatically generated by scanning a folder or folder tree. All files with matching characteristics are grouped together, based on the information in their FITS headers. Alternatively, the user can manually create groups.

All the files in a group are combined together using an average, median, sigma combine, or SD Mask algorithm to make an internal "master frame" that is used to calibrate images. This master frame is not visible to the user.

Some users like to generate and save master frames as files, so they can be quickly reloaded at a later date. The Set Calibration command allows you to do this at the click of a button. Note however that this step is recommended, but completely optional and is not required for calibration to work.

When master frames are generated and saved to disk, the list of groups is automatically replaced by the master frame files. If bias subtraction is enabled and suitable groups are available, the darks and flat masters will automatically be bias-subtracted. Similarly the flat masters will be dark subtracted if dark subtraction is enabled and suitable groups are available. When the masters are saved to disk, they are tagged with header flags to indicate whether these subtractions have been done. That way when they are loaded again later, MaxIm DL will know whether they need to have these subtractions performed or not.

When MaxIm DL is used to calibrate an image, all available groups are scanned to determine compatibility. The following image parameters are used (if available), as extracted from the image header:

Groups are rejected if either the pixel size or binning does not match exactly. If the image is a subframe it also has to fit within the calibration frame.

If an exact match is not found, then for bias and dark frames, a matching temperature setpoint is the highest priority. Failing an exact match, the closest temperature is used (tip: if an exact temperature match is not available, you may want to turn on Auto Optimize scaling mode).

For dark frames, it then looks for the same exposure time; failing that, it finds the closest exposure time. It is strongly recommended that the Auto Scale mode always be turned on, so that the closest matching dark frame can be scaled if an exact match is not available.

For flat-fields, if a matching filter band is found, then that band is used. If no matching filter band is found, then it looks for a generic "FLAT" group with no filter band (usually indicating an unfiltered flat).

If more than one matching group is found, then the first one in the list is used. Normally this doesn't happen if the groups are auto-generated, but it may happen for manually added groups.

Using this "calibration groups" approach, MaxIm DL is able to provide the following advanced  capabilities, such as: