1 September 2017 Optimization of pyKLIP's forward model matched filter for the GPI Exoplanet Survey
Author Affiliations +
The Gemini Planet Imager (GPI) is an adaptive optics instrument for direct imaging of young self-luminous extrasolar planets in the immediate vicinity of bright stars. Such high contrast images are often dominated by speckle artifacts, which need to be removed with sophisticated data processing. Ref. 1 introduced a new matched filter for direct detection of point sources using the KLIP speckle subtraction algorithm including a forward model of the planet PSF. We present here a much more thorough analysis of the free parameters of the implementation of this algorithm in the Python git repository pyKLIP. We perform a series of one dimensional optimization over pyKLIP parameters: the reference library exclusion criterion, the number of principal components, and the size of the sectors. We also study the effect of a spectral template mismatch between the real planets and the reduction spectrum in four GPI spectral bands (J, H, K1, K2).
© (2017) COPYRIGHT Society of Photo-Optical Instrumentation Engineers (SPIE). Downloading of the abstract is permitted for personal use only.
Jean-Baptiste Ruffio, Bruce Macintosh, Jason J. Wang, Laurent Pueyo, "Optimization of pyKLIP's forward model matched filter for the GPI Exoplanet Survey", Proc. SPIE 10400, Techniques and Instrumentation for Detection of Exoplanets VIII, 1040027 (1 September 2017); doi: 10.1117/12.2274539; https://doi.org/10.1117/12.2274539

Back to Top