As the quantity of astronomical data available continues to exceed the resources available for analysis, recent advances in artificial intelligence encourage the development of automated classification tools. This project, my undergraduate Honors Capstone Project at Seattle Pacific University, develops a framework for training a deep neural network capable of classifying individual astronomical images by describing techniques to extract and label these objects from large images.
The deep neural network does not rely on any particular source of data but to obtain labeled images I used Astrometry and processed them with Source Extractor to produce training data.
Astrometry is of tremendous use for this purpose but the methodologies described herein for processing FITS files with Source Extractor and AstroPy are relevant regardless of image source.
This repository accompanies the academic paper Using Deep Neural Networks to Classify Astronomical Images by Andrew Macpherson. The fully accompanying paper is published in the Seattle Pacific University Digitial Commons and is protected under copyright with educational use permitted. For additional usage, please contact the author.
This repository is designed for programmers and is organized in the following manner:
📁Appendixes - Contains appendixes that are not given in the paper, namely Appendixes C, D, & E. These contain basic information for objects in the New General Catalog, Henry Draper Catalog, and Index Catalog.
📁Data - Contains sample raw astronomical images in FITS format. These sample files were used to create labeled images
📁Images - Contains labeled cutouts from sample FITS files.
📁config - Contains configuration files for Source Extractor, modified as described in the paper to adjust the appropriate parameters and extract relevant data.
This material was presented at the following university conferences. Each was recorded in some capacity and is available for viewing online.
⭐ Award Winner - Top Oral Presentation ⭐
21st Annual Erickson Conference
Seattle Pacific University
May 5th, 2023
Video Link
26th Annual Undergraduate Research Symposium
University of Washington
May 19th, 2023
2nd Annual Honors Research Symposium
Seattle Pacific University
May 20th, 2023
Video Link
Should you have any questions or inquiries about this project, please reach out to me via one of the following contact methods:
Portfolio: andrewm.tech
LinkedIn: linkedin.com/in/macphersona
Email: [email protected]