batch processing master flat field question

[whwang]

Hi,

I have a very general question regarding how the master flat is generated in batch processing. 

We usually normalize the average value of a flat field image to 1.0, or whatever value.  When PI's batch processing script combine various flat images to form a master flat, does it first normalize all flat images to 1.0 (or a specific value) and create the master flat?  Or it directly combines the multiple flat exposures to form a master flat and then normalize the master flat?

Cheers,
Wei-Hao

[RickS]

We usually normalize the average value of a flat field image to 1.0, or whatever value.  When PI's batch processing script combine various flat images to form a master flat, does it first normalize all flat images to 1.0 (or a specific value) and create the master flat?  Or it directly combines the multiple flat exposures to form a master flat and then normalize the master flat?

Normalization occurs twice during image integration, once to prepare each sub for rejection and again prior to combining the data for output.  For flat frames, the appropriate normalization options are "Equalize Fluxes" for rejection and "Multiplicative" for the final integration.  These are the options that BPP uses when building a master flat.  All normalization occurs prior to the final integration.

This is described well in the documentation for ImageIntegration if you want to dig into the details.

Cheers,
Rick.

[whwang]

Hi Rick,

The normalization schemes you mentioned make sense to me.  I will check the document for the details.

Thanks.