This article was accepted into the corpus but its outbound wikilinks were never NER-processed — typical at the deepest BFS hop or when the run's entity cap was reached. No expansion funnel to show.
| DAOStarFinder | |
|---|---|
| Name | DAOStarFinder |
| Developer | Astropy Community |
| Released | 2018 |
| Programming language | Python |
| Repository | GitHub |
| License | BSD |
| Website | Astropy Project |
DAOStarFinder DAOStarFinder is a source detection algorithm adapted for modern astronomical imaging that identifies point-like sources in two-dimensional detector arrays. It is an implementation inspired by classical point-spread function fitting and peak-finding routines used in stellar photometry, integrating ideas from historical techniques and contemporary software ecosystems. The routine is widely used in survey pipelines, space-mission data reduction, and ground-based observatory processing.
DAOStarFinder traces conceptual roots to the DAOPHOT family of algorithms developed for crowded-field photometry at observatories such as Kitt Peak National Observatory, Cerro Tololo Inter-American Observatory, Palomar Observatory, and missions like Hubble Space Telescope and Kepler. It is implemented within the Astropy ecosystem alongside affiliated packages maintained by the Astropy Project contributors, and interacts with libraries including NumPy, SciPy, Matplotlib, and Pandas for data manipulation and visualization. The design aims to balance heritage from methods used by astronomers at institutions such as Harvard–Smithsonian Center for Astrophysics, Max Planck Institute for Astronomy, and European Southern Observatory with modern software engineering practices from platforms like GitHub and Bitbucket.
DAOStarFinder applies a convolution-based peak detection using a Gaussian-like kernel, then fits local maxima with a model approximation to estimate centroids and fluxes. The approach parallels fitting strategies found in the literature from groups affiliated with Space Telescope Science Institute, Caltech, and Carnegie Institution for Science. Internally it employs routines provided by SciPy optimization and signal modules, array operations from NumPy, and masking handled in patterns used by Pandas and Dask for scalable workloads. Implementation choices reflect contributions by developers connected to projects such as SunPy, gammapy, and photutils maintainers who coordinate through community forums like NumFOCUS and conferences including Astronomical Data Analysis Software and Systems.
Performance characteristics depend on image size, crowding, background characteristics, and detector properties found in instruments from facilities like Very Large Telescope, Subaru Telescope, James Webb Space Telescope, and Sloan Digital Sky Survey. Accuracy is typically benchmarked against synthetic fields and catalogs generated with tools used by teams at European Space Agency, NASA, and survey projects such as Pan-STARRS and LSST (now Vera C. Rubin Observatory). Statistical comparisons often reference metrics and procedures employed by groups at Carnegie Mellon University, University of Cambridge, and Princeton University that use cross-matching, false-positive rates, and astrometric residuals. Profiling studies utilize performance tooling popularized by developers at Google, Microsoft Research, and Intel to assess computational scaling, memory footprint, and parallelization behavior.
DAOStarFinder is exposed to users through the Photutils package, an affiliated package of Astropy Project, making it accessible to pipelines used by teams at Space Telescope Science Institute, Max Planck Institute for Astronomy, and survey collaborations like Dark Energy Survey. Typical workflows integrate DAOStarFinder with source catalog construction methods developed by groups at University of California, Berkeley, University of Arizona, and Yale University, and combine results with photometric tools from SExtractor-inspired pipelines used in projects at IAC and NOAO. The API follows conventions consistent with other Astropy subpackages and interoperates with table structures familiar to users of Astropy Table, enabling downstream analysis by researchers at institutions such as University of Oxford, Columbia University, and University of Toronto.
DAOStarFinder is compared with classical packages and modern alternatives including SExtractor, DAOPhot (original implementations at Dominion Astrophysical Observatory and applied at Mount Wilson Observatory), PSF-fitting codes employed by teams at Space Telescope Science Institute, and newer machine-learning approaches from groups at MIT, Stanford University, and Carnegie Mellon University. Other alternatives include crowded-field photometry suites developed in collaborations like OGLE and MOA, and deep-learning detectors trained by consortia such as Zwicky Transient Facility and Pan-STARRS. Trade-offs typically involve detection completeness, processing speed, robustness to background structure seen in data from Chandra X-ray Observatory and XMM-Newton, and ease of integration into pipelines run by observatories such as Keck Observatory.
DAOStarFinder has been used in studies ranging from variable-star searches at Harvard College Observatory and exoplanet transit follow-up coordinated with Kepler and TESS teams, to transient discovery programs operated by Zwicky Transient Facility and ASAS-SN. It appears in image-processing chains for survey data from Sloan Digital Sky Survey, Pan-STARRS, and calibration tasks for instruments on James Webb Space Telescope and Hubble Space Telescope. Example use-cases include crowded-cluster photometry pursued by research groups at European Southern Observatory, proper-motion work by groups at Carnegie Institution for Science, and pipeline validation in collaborations affiliated with National Optical-Infrared Astronomy Research Laboratory.
Category:Astronomical software