IMAGE RECONSTRUCTION USING ARTIFICIAL INTELLIGENCE (AI) MODULES AGNOSTIC TO IMAGE ACQUISITION SETTINGS

摘要

An image processing system for determining an image of a scene using AI modules that are agnostic to image acquisition settings. The system comprises a processor that executes an image acquisition settings estimator module to estimate a forward operator using setting data. Further, recursively estimates the image of the scene until a termination condition is met, where for a current iteration, the processor executes an image acquisition inverter module with the estimated forward operator and a previous estimate of the image of the scene estimated during a previous iteration to produce the current noisy image, executes a noise level estimator module with the previous estimate of the image to produce a current noise level, and executes a denoiser to remove the current noise level from the current noisy image to produce a current image.