A Machine Learning Approach to the Detection of Ghosting and Scattered Light Artifacts in Dark Energy Survey Images
Authors:
Chihway Chang,
Alex Drlica-Wagner,
Stephen M. Kent,
Brian Nord,
Donah Michelle Wang,
Michael H. L. S. Wang
Abstract:
Astronomical images are often plagued by unwanted artifacts that arise from a number of sources including imperfect optics, faulty image sensors, cosmic ray hits, and even airplanes and artificial satellites. Spurious reflections (known as "ghosts") and the scattering of light off the surfaces of a camera and/or telescope are particularly difficult to avoid. Detecting ghosts and scattered light ef…
▽ More
Astronomical images are often plagued by unwanted artifacts that arise from a number of sources including imperfect optics, faulty image sensors, cosmic ray hits, and even airplanes and artificial satellites. Spurious reflections (known as "ghosts") and the scattering of light off the surfaces of a camera and/or telescope are particularly difficult to avoid. Detecting ghosts and scattered light efficiently in large cosmological surveys that will acquire petabytes of data can be a daunting task. In this paper, we use data from the Dark Energy Survey to develop, train, and validate a machine learning model to detect ghosts and scattered light using convolutional neural networks. The model architecture and training procedure is discussed in detail, and the performance on the training and validation set is presented. Testing is performed on data and results are compared with those from a ray-tracing algorithm. As a proof of principle, we have shown that our method is promising for the Rubin Observatory and beyond.
△ Less
Submitted 21 May, 2021;
originally announced May 2021.
