Improving spatial accuracy and signal-to-noise ratio through Babinet-principle-based coherent amplitude modulation imaging.

A Babinet-principle-based coherent amplitude modulation imaging (BP-CAMI) method is proposed for accurate illumination characterization. It employs a random amplitude plate and a complementary constraint derived from the Babinet principle, establishing a dual-constraint framework that provides algorithmic rather than hardware redundancy. This approach effectively suppresses noise and artifacts during phase retrieval. Simulations and experiments on biomedical samples, resolution targets, and phase objects confirm substantial improvements in signal-to-noise ratio and spatial accuracy over conventional methods. Notably, BP-CAMI recovers richer high-frequency details than the ptychographic iterative engine, with quantitative gains of 2.85 dB PSNR and 7.81 μm resolvable linewidth, yet requires no hardware modifications. The work offers a practical, algorithm-driven route to enhance performance across coded-imaging systems.