deepCR: Cosmic Ray Rejection with Deep Learning

Abstract

Cosmic ray (CR) identification and replacement are critical components of imaging and spectroscopic reduction pipelines involving solid-state detectors. We present deepCR, a deep learning based framework for CR identification and subsequent image inpainting based on the predicted CR mask. To demonstrate the effectiveness of this framework, we train and evaluate models on Hubble Space Telescope ACS/WFC images of sparse extragalactic fields, globular clusters, and resolved galaxies. We demonstrate that at a false positive rate of 0.5%, deepCR achieves close to 100% detection rates in both extragalactic and globular cluster fields, and 91% in resolved galaxy fields, which is a significant improvement over the current state-of-the-art method LACosmic. Compared to a multicore CPU implementation of LACosmic, deepCR CR mask predictions run up to 6.5 times faster on CPU and 90 times faster on a single GPU. For image inpainting, the mean squared errors of deepCR predictions are 20 times lower in globular cluster fields, 5 times lower in resolved galaxy fields, and 2.5 times lower in extragalactic fields, compared to the best performing non-neural technique tested. We present our framework and the trained models as an open-source Python project, with a simple-to-use API. To facilitate reproducibility of the results we also provide a benchmarking codebase.

Accepted for publication in ApJ. 12 pages, 6 figures. An open-source Python package, deepCR, which implements the approach in this paper is at https://github.com/profjsb/deepCR. Figures and benchmarks can be reproduced at https://github.com/kmzzhang/deepCR-paper. Blog post describing this work is at https://medium.com/@kemingzhang/deepcr-2fa610655655

References in corpus (3)

• The NumPy array: a structure for efficient numerical computation
• scikit-image: Image processing in Python
• Binary companions of evolved stars in APOGEE DR14: Search method and catalog of ~5,000 companions

Cited by in corpus (17)

• The electron-capture origin of supernova 2018zd
• Euclid: Early Release Observations -- Programme overview and pipeline for compact- and diffuse-emission photometry
• Project Dinos I: A joint lensing-dynamics constraint on the deviation from the power law in the mass profile of massive ellipticals
• Beyond Gaussian Noise: A Generalized Approach to Likelihood Analysis with non-Gaussian Noise
• Classification of Periodic Variable Stars with Novel Cyclic-Permutation Invariant Neural Networks
• The PAU survey: Estimating galaxy photometry with deep learning
• The PHANGS-HST-Halpha Survey: Warm Ionized Gas Physics at High Angular resolution in Nearby GalaxieS with the Hubble Space Telescope
• deepSIP: Linking Type Ia Supernova Spectra to Photometric Quantities with Deep Learning
• Likelihood-Based Jump Detection and Cosmic Ray Rejection for Detectors Read Out Up-the-Ramp
• Enhanced Astronomical Source Classification with Integration of Attention Mechanisms and Vision Transformers
• Cleaning Images with Gaussian Process Regression
• A Machine Learning Approach to the Detection of Ghosting and Scattered Light Artifacts in Dark Energy Survey Images
• Exploring the detection of AQNs in large liquid detectors
• Cosmic Ray Rejection with Attention Augmented Deep Learning
• Deep Learning the Forecast of Galactic Cosmic-Ray Spectra
• Combined JWST and HST Deep Imaging to Characterize the Smallest Known Trans-Neptunian Objects
• Can AI Dream of Unseen Galaxies? Conditional Diffusion Model for Galaxy Morphology Augmentation