Method Removes Starlight to Reveal Faint Exoplanets
A Northwestern-led team developed a physics-based framework that subtracts starlight from telescope images to make distant planets easier to see.
Updated on Nov. 4, 2026 in Space
Researchers at Northwestern University developed a method that removes starlight from telescope images. In tests on James Webb Space Telescope data, it improved sensitivity to faint exoplanets by two to 15 times over conventional techniques.
Why it matters
Host stars' glare can overwhelm the faint light of exoplanets, making direct imaging difficult. Existing techniques leave residual glare when tiny changes in telescope optics occur, so a better subtraction method expands what astronomers can detect.
Tests on Webb's NIRCam observations showed sensitivity gains of two to 15 times over conventional methods, and the method produced the first image of the known giant planet 51 Eridani b at 4.3 micrometers where conventional processing had failed. Whether the framework will reach comparable gains on Earth-like planets has not yet been demonstrated.
The players
Northwestern University
A private research university in Evanston, Illinois, known for its work across engineering, physics, and astronomy.
Rodrigo Ferrer-Chavez
The Northwestern researcher who led the team that developed the FLIP starlight-subtraction framework.
The details
The framework, called Physics-based Learning for Imaging Planets (FLIP), was developed by a team led by Rodrigo Ferrer-Chavez. It models how light passes through the telescope, adjusts for optical imperfections, and simulates the glare to be subtracted, combining a physics-based telescope model with optimization techniques used in machine learning.
Big Picture
The method extends what the James Webb Space Telescope's Near-Infrared Camera can reveal by reprocessing its existing observations, reaching planets that standard starlight subtraction missed. The team hopes continued development will move direct imaging of Earth-like planets closer to reality.
Researchers hope further development and testing will bring direct images of Earth-like planets closer, and the framework can be extended to other current and future ground-based and space-based observatories. Reducing computational demands for larger ground-based datasets is also on the team's list of next steps.
The takeaway
Better processing of existing telescope data can unlock planets that were previously hidden in the glare of their host stars. Techniques like FLIP show that refining how astronomers model their instruments may matter as much as building new ones.
Further reading
Explore more Space coverage to see how astronomers are pushing telescope imaging further.
