The Lucy–Richardson–Rosen algorithm is widely used for image restoration, but suffers from slow convergence or failure when analyzing images with severe optical aberrations and high noise. To address these limitations, we propose the Differential Lucy–Richardson–Rosen algorithm which enhances both robustness and convergence speed. By integrating a Hartmann–Shack wavefront sensor into the imaging system, our proposed algorithm directly measures wavefront distortions to accurately estimate the spatially varying point spread function, enabling high-fidelity non-blind deconvolution, even for images acquired by ground-based telescopes, with significant optical imperfections. Extensive simulations and experiments demonstrate that our proposed algorithm outperforms its predecessor in image quality and computational efficiency under challenging aberration and noise conditions. Its rapid and stable performance makes it particularly suitable for real-time or near-real-time astronomical imaging, where reliable, high-resolution recovery is critical. This work advances computational imaging for next-generation astronomical instrumentation through a tightly coupled hardware–algorithm framework.
Zheng et al. (Sun,) studied this question.