Coherent differential imaging: a new image processing techniques towards the first image of a habitable zone exoplanet

Direct imaging of exoplanets represents one of the most thrilling and rapidly advancing domains in contemporary astronomy. This method enables the study of exoplanetary atmospheres by capturing light directly emitted or reflected by these distant worlds. While current technology already allows the characterization of large gas giants, the ultimate ambition of direct imaging is to extend this capability to Earth-like planets. Such an achievement would enable the measurement of atmospheric compositions and even the search for potential biosignatures, marking a monumental leap in our understanding of planetary systems beyond our own. However, the extreme contrast between the brightness of host stars and the faint light of their orbiting planets poses a formidable instrumental challenge. Detecting an exoplanet in this context is comparable to spotting a firefly next to a lighthouse from thousands of kilometers away.
Coronagraphs are sophisticated optical instruments specifically designed to address this challenge. Their primary function is to suppress the overwhelming starlight, thereby allowing the detection of the much fainter exoplanet signal. The next generation of giant telescopes, including the European Extremely Large Telescope (ELT) in Chile and NASA’s Habitable Worlds Observatory, will heavily rely on these advanced coronagraphic techniques. These telescopes aim to image Earth-like planets located within the habitable zones of nearby stars. The Planetary Camera and Spectrograph (PCS), a second-generation coronagraphic instrument planned for the ELT, is specifically designed to explore the habitable zones around red dwarf stars, where rocky planets could potentially exist. Meanwhile, HWO will focus on detecting habitable-zone planets around stars similar to our Sun. Both of these ambitious missions require the development of new coronagraphic techniques to achieve the extreme levels of star light suppression necessary to detect these faint and distant worlds.The ECHOES project, led by Johan Mazoyer and funded by the European Research Council (ERC), aims to significantly advance the field of exoplanet imaging by developing innovative coronagraphic techniques tailored for these missions. This PhD opportunity is fully funded as part of the ECHOES ERC project.
 

The primary objective of this PhD is to develop and validate innovative post-processing techniques for the detection of exoplanets in coronagraphic images. Unlike traditional methods that rely on angular or spectral diversity to differentiate planetary and stellar light, this project is to develop Coherent Differential Imaging (CDI) to exploit the fundamental property of light coherence. The PhD will focus on two complementary approaches: 1) Enhancing CDI with machine learning: improve this technique using convolutional neural networks (CNNs) trained on simulated data, enabling faster and more robust planet detection. 2) Developing a new CDI method using another modulation technique. This PhD will explore how to adapt this technique for single-frame model free post-processing method.

The PhD will employ a combination of simulation and experimental validation. First, use and develop existing coronagraphic simulation tools in python to develop innovative algorithms, then conduct tests on the THD2 testbed to validate the algorithm’s performance under realistic conditions, and finaly participate in observing runs at the Very Large Telescope in Chile to test the algorithm with the SPHERE+ instrument.

The successful candidate will be part of the exoplanet team at LIRA / Paris Observatory - PSL. Our team, and Dr. Mazoyer in particular, is committed to fostering a diverse and dynamic work environment by actively recruiting individuals of all genders and nationalities. In Paris observatory, arguably the world oldest institution for astronomical research, our team is one of the largest and most dynamic exoplanet research groups in Europe, in particular for direct imaging. The project provides a collaborative network, engaging with leading experts in optics, astrophysics, and machine learning from institutions such as ESO, NASA, and ONERA. This PhD in astronomical instrumentation for future space missions sits at the intersection of engineering, and astrophysics, a uniquely interdisciplinary opportunity in a very promising field. It is a great opportunity for applicants interested in starting a high impact academic career but also open exciting job prospects in space agencies or industries, in optics, aerospace, or advanced high-tech instrumentation.

Paris Observatory Exoplanet’s team: https://lira.obspm.fr/-Pole-Systemes-Exoplanetaires-?lang=en (please contact our current and former PhDs to learn about our work environment!)

ECHOES is an interdisciplinary project which values diverse expertise: we welcome applicants from a wide range of backgrounds, even if you never studied astrophysics previously! Candidates should hold a Master’s degree in Physics or Astrophysics, Optics, or computer science, with an interest in instrumentation, experimental work or signal processing. Experience with programming (Python) is essential. The candidate will work in a collaborative, interdisciplinary and international environment: fluency in English, both written and spoken, is required for effective scientific communication.