How a free AI tool is fixing NASA's JWST images
Two PhD students from the University of Sydney, Louis Desdoigts and Max Charles, built an AI tool called AMIGO that fixed blurry images from the James Webb Space Telescope—all without touching a single piece of hardware.
Their software tackled a tricky problem known as the "brighter-fatter effect," where electric charge messes with image pixels and causes distortion.
AMIGO uses neural networks to model JWST's optics and electronics
AMIGO uses neural networks to model JWST's optics and electronics, letting it spot and digitally repair distortions remotely.
As Professor Tuthill put it, instead of sending astronauts for repairs, these students "fixed things with code"—a brilliant example of how Australian innovation can make a global impact in space science.
With AMIGO on board, JWST can now snap much clearer photos
With AMIGO on board, JWST can now snap much clearer photos—like spotting a faint exoplanet or getting detailed views of Jupiter's volcanic moon Io.
It's a win for science (and budgets), showing how smart software is making space exploration sharper and more efficient.