Skip navigation
Please use this identifier to cite or link to this item: http://arks.princeton.edu/ark:/88435/dsp01bk128d296
 Title: Planet identification for speckle-limited coronagraphic images Authors: Young, Elizabeth Jensen Advisors: Kasdin, N. J Contributors: Mechanical and Aerospace Engineering Department Keywords: CoronagraphsDirect ImagingExoplanetsHigh Contrast ImagingSpeckles Subjects: Aerospace engineeringAstronomy Issue Date: 2015 Publisher: Princeton, NJ : Princeton University Abstract: The first successful direct image of an exoplanet was taken in 2008. This field is rapidly advancing in a quest to image Earth-like planets and look for signs of life. All direct imaging techniques use a coronagraph to block the bright starlight and reveal faint companions nearby. A significant challenge in the field is the fact that quasi-static speckles and planets appear to be similar within an image. However, speckles are derived from a single coherent source (the star), and are incoherent with the light from a planet. Therefore, the speckle pattern in an image can be changed, while leaving any planet light unaffected, by moving a deformable mirror within the coronagraphic system. This dissertation presents a technique to analyze a series of images containing different speckle patterns in order to confidently identify planets. One key difficulty in using existing techniques to analyze such images is that they rely on having a known or knowable background in order to extract the planet(s). Instead, we focus on simultaneously estimating the unknown planet and background intensities. Using these estimates, we present a Bayesian analysis in order to assess the evidence towards the existence of a planet. We explore three approaches to calculate the required estimates, namely stacking all images together, estimating the parameters in every image individually, and concurrently estimating a single planet intensity but multiple background intensities for each image. For each approach, we demonstrate how to select a detection threshold, number of images, and overall integration time to match desired probabilities of false alarms and missed detections. We compare the three approaches using both theoretical analysis and simulation results. We find that stacking images provides the greatest planet-detection capability. However, the approach of concurrently estimating the background in each image shows the potential to be improved through technical modifications, and already performs similarly to stacking in some cases. Finally, we use laboratory results to demonstrate that speckle patterns can be changed in the ways that we assumed. URI: http://arks.princeton.edu/ark:/88435/dsp01bk128d296 Alternate format: The Mudd Manuscript Library retains one bound copy of each dissertation. Search for these copies in the library's main catalog: http://catalog.princeton.edu/ Type of Material: Academic dissertations (Ph.D.) Language: en Appears in Collections: Mechanical and Aerospace Engineering

Files in This Item:
This content is embargoed until 2017-11-24. For more information contact the Mudd Manuscript Library.

Items in Dataspace are protected by copyright, with all rights reserved, unless otherwise indicated.