About the generation of flat for OSC ,
what is the advantage or difference of CCDStack to
Ron Wadaski's 'color flat maker' MaxIm plug-in.

Regards,

Noboru

I am not familiar with that plug-in, nor have I used MaxIm in the last several years.

Here are a few considerations for Flat Fielding to work effectively:

The non-photonic bias (including pedestal) must be subtracted from the image and the flat prior to the actual flat-divide. Many programs add a pedestal (usually 100 ADU) to “dark subtracted” images and that pedestal must be removed. The original reason for the pedestal was that earlier (and some current) software used unsigned 16 bit data structures, which cannot handle negative numbers, so the 100 ADU boost is intended to lift most pixels above zero. CCDStack does not use a pedestal (or pedestal = 0) because it stores the data as 32 bit floating point and allows negative numbers. If you are importing a dark subtracted image from other software, it is important to enter the value of that pedestal.

Bayer images should be divided into 4 sub-images based on the Bayer Matrix position then each of the sub-images should be flatted independently. If this is not done then the flat can impart color distortions onto the image (and might even retain some of the blocky matrix). CCDStack does this when the image is identified as “Bayer” (via Camera Manager setting or FITS header or DNG metadata).

Note that “Black box” software may or may not correctly handle these issues and it can be difficult for the user to know what is really happening. CCDStack gives control of these issues to the user (e.g. for the Calibration box accommodates Flat dark/bias subtraction or uses the pedestal). Of course, some users don’t know or care what’s going on and CCDStack is designed so that these operations will likely be correct if the user simply follows standard procedures within CCDStack. But users who try to mix-and-match software do so at their own risk (e.g. using MaxIm for part of the calibration and CCDStack for another part can result in errors).

Stan

This message has been edited. Last edited by: Stan,