find-asteroids

Find moving objects in detection catalogs.

Installation via PyPI:

$ python -m pip install find-asteroids

Usage (docs/notebooks/search.ipynb):

usage: find-asteroids [-h] --catalog CATALOG [--psfs PSFS] --velocity VELOCITY VELOCITY --angle ANGLE ANGLE --dx DX --num-results
                      NUM_RESULTS --results-dir RESULTS_DIR [--precompute] [--gpu] [--gpu-kernels] [--device DEVICE]
                      [--output-format OUTPUT_FORMAT] [--refine-iterations REFINE_ITERATIONS]

optional arguments:
  -h, --help            show this help message and exit
  --catalog CATALOG     The detection catalog to search. An astropy-readable table containing at least 'ra', 'dec', and 'time' columns
                        (with units). (default: None)
  --psfs PSFS           An astropy-readable table containing a 'psf' column (with units) that specifies the PSF-widths of the images from
                        which the detection catalog is derived. If not provided, a value of 1 arcsec is assumed. (default: None)
  --velocity VELOCITY VELOCITY
                        The velocity range over which to search, in units of deg/day. (default: None)
  --angle ANGLE ANGLE   The on-sky angles over which to search, in units of deg. (default: None)
  --dx DX               Search bin-width, in units of the PSF-width. (default: None)
  --num-results NUM_RESULTS
                        Number of results to produce. (default: None)
  --results-dir RESULTS_DIR
                        The directory into which to write results. (default: None)
  --precompute          Precompute projected positions of detections for all trial velocities (uses more memory, but may be faster).
                        (default: False)
  --gpu                 Run the core-search components of the algorithm on GPU. (default: False)
  --gpu-kernels         Run the entirety of the search algorithm on the GPU. (default: False)
  --device DEVICE       The GPU device number to use. (default: -1)
  --output-format OUTPUT_FORMAT
                        The astropy.table supported format for writing results. (default: ecsv)
  --refine-iterations REFINE_ITERATIONS
                        The number of times to refine a candidate result. (default: 1)

References:

• Stetzler, S. et al. (2025) An Efficient Shift-and-Stack Algorithm Applied to Detection Catalogs; arXiv:2509.26279