Detection and Characterization of Exoplanets and Disks using Projections on Karhunen-Loeve Eigenimages

TL;DR

This paper introduces a PCA-based method for high-contrast imaging that improves PSF subtraction robustness and enables accurate characterization of exoplanets and disks, applicable to space and ground-based data.

Contribution

The paper presents a novel PCA-based PSF subtraction technique that enhances detection robustness and allows forward modeling for better source characterization.

Findings

Comparable detection performance to LOCI with increased robustness

Enables forward modeling for accurate source characterization

Applicable to HST, ground-based ADI, and IFS data

Abstract

We describe a new method to achieve point spread function (PSF) subtractions for high- contrast imaging using Principal Component Analysis (PCA) that is applicable to both point sources or extended objects (disks). Assuming a library of reference PSFs, a Karhunen-Lo`eve transform of theses references is used to create an orthogonal basis of eigenimages, on which the science target is projected. For detection this approach provides comparable suppression to the Locally Optimized Combination of Images (LOCI) algorithm, albeit with increased robustness to the algorithm parameters and speed enhancement. For characterization of detected sources the method enables forward modeling of astrophysical sources. This alleviates the biases in the astrometry and photometry of discovered faint sources, which are usually associated with LOCI- based PSF subtractions schemes. We illustrate the algorithm performance using archival Hubble Space Telescope (HST) images, but the approach may also be considered for ground-based data acquired with Angular Differential Imaging (ADI) or integral-field spectrographs (IFS).