Quickstart

Directory layout

LBCgo expects raw LBC FITS files in a raw/ subdirectory of your working directory. Files should follow the LBT naming convention lbc[b|r].YYYYMMDD.NNNNNN.fits. Calibration frames (biases, sky flats) and science frames can all live together in raw/; the pipeline sorts them by IMAGETYP header keyword.

Expected layout before running:

my_night/
└── raw/
    ├── lbcb.20230101.000001.fits   # LBC-Blue science frame
    ├── lbcb.20230101.000002.fits
    ├── lbcr.20230101.000010.fits   # LBC-Red science frame
    └── ...

Standard run

For a typical night where all calibration data are present, reduce everything in one call:

from LBCgo.lbcproc import lbcgo
lbcgo()

This processes both LBC-Blue and LBC-Red data for all filters found in ./raw/, applies overscan correction, master flat fields, extracts individual chips, and performs astrometric registration against GAIA-DR3.

Output files are organised into subdirectories by target name and filter, e.g. NGC891/g-SLOAN/.

Common options

Process a single filter only:

lbcgo(filter_names=['g-SLOAN'])

Skip astrometry (useful for a quick first look, or when the external tools are not installed):

lbcgo(do_astrometry=False)

LBC-Blue data only:

lbcgo(lbcb=True, lbcr=False)

Missing or disabled chips (e.g., chip 3 was offline):

lbcgo(lbc_chips=[1, 2, 4])

Apply bias frames (not done by default because LBC bias levels are stable and well-characterised by the overscan):

lbcgo(bias_proc=True)

Multi-filter example (both cameras)

To reduce SDSS r-band from LBC-Red and g-band from LBC-Blue separately:

from LBCgo.lbcproc import lbcgo

# Red camera: r-band
lbcgo(lbcr=True, lbcb=False, filter_names=['r-SLOAN'])

# Blue camera: g and U
lbcgo(lbcr=False, lbcb=True, filter_names=['g-SLOAN', 'SDT_Uspec'])

Running astrometry separately

The astrometric step can be deferred or re-run independently using go_register():

from glob import glob
from LBCgo.lbcregister import go_register

# Run full astrometric pipeline on previously reduced data
fltr_dirs = glob('NGC891/*/')
go_register(fltr_dirs)

Run the steps individually if you need to inspect intermediate results or re-run just SCAMP and SWarp:

# Step 1: Source extraction only
go_register(fltr_dirs, do_sextractor=True, do_scamp=False, do_swarp=False)

# Step 2: Astrometric calibration (more iterations for precision)
go_register(fltr_dirs, do_sextractor=False, do_scamp=True, do_swarp=False,
            scamp_iterations=5)

# Step 3: Image co-addition (after removing any frames with bad solutions)
go_register(fltr_dirs, do_sextractor=False, do_scamp=False, do_swarp=True)

Output products

After a full reduction run, the working directory contains:

  • <target>/<filter>/ — flat-fielded individual chip images (*_1.fits … *_4.fits)

  • <target>/<filter>/<object>.<filter>.mos.fits — astrometrically registered co-add (if do_astrometry=True)

  • <target>/<filter>/<object>.<filter>.mos.weight.fits — corresponding weight map

  • data/ — intermediate files (overscan-corrected, flat-fielded MEF images)