CROCOA

CRoss COrrelation Alignment tool.

Software package for using 2D crosscorrelation to align astronomical images.

Requirements

The primary requirements are astropy and scipy in order to manage the fits files and 2d correlation function. Additionally drizzlepac is also used to drizzle the frames in order to account for differing pixel grids and rotation

The package is tested on python 3.7

Installation

For having an editable installation activate your preferred virtualenv and run the following in the CROCOA main directory

$ python setup.py develop

Usage

Example:

import os
import glob
from crocoa.imalign import align
from crocoa.drizzle_config import config

# Review the drizzle config settings
for k in config.keywords():
    print(k, ': ', config[k])

# Set the Ra and dec in the drizzle config to accurate values
config['final_ra'] = 1.12637
config['final_dec'] = -10.19156

# Prepare your directories
source_images = glob.glob("SDSSJ0004-1011/F150LP/*")

destination_dir = "SDSSJ0004-1011/F150LP_aligned/"
os.mkdir(destination_dir)

# Run alignment
align(source_images, destination_dir, config)

Tips and tricks

Image size

The 2D correlation is very sensitive to the relative size of the target to the frame i.e. a small target in a large frame can give poor results. Therefore it can be useful to minimize the size of the frame considered while still makeing sure that the full target is contained.

In cases where the alignment is very poor this can require manual correction of the coordinate system beforehand.

Image normalization

How the image is normalized before being run through the 2D correlation can also impact the quality of the alignment. test setting the normalization keyword to different settings.

Always check manually

What the title says. The system is most definitely not infallible.

Contributions

Contributions are welcome in the form of pull requests and issue submissions.