Mémoire

Abstract

To this day, thousands of exoplanets have been discovered thanks to various detection methods. With such a population of exoplanets starting to become statistically significant, it is now possible to try to infer general properties of planetary systems and study how frequent each configuration is. In order to truly understand this great diversity in observed exoplanets, studying planet formation is a necessary and crucial step; however, there are still many remaining mysteries related to the birth of planetary systems. To blame are the observations, or lack thereof, of these processes in action. Direct imaging is the only detection method able to study these protoplanetary disks in the act of forming planets. However, direct imaging also presents many challenges related to the proximity and brightness of the star around which the observations are conducted. State-of-the-art instruments, as well as efficient post-processing algorithms, are necessary to suppress the starlight as much as possible in order to reveal the faint signals around the star. The goal of this project is to focus on the post-processing phase and analyze multiple direct imaging datasets of protoplanetary disks, the birth environment of planets, in an attempt to detect planetary embryos, otherwise called protoplanets. To this end, various post-processing algorithms will be compared, and additional new ones will be developed if the need arises. If no protoplanet is found, contrast curves will be calculated in order to place stringent constraints on the existence of hypothetical undetected signals.