Pyramidal lens enables mesoscale structured illumination by surpassing the spatial bandwidth limit of spatial light modulators

Structured illumination has emerged as a pivotal technique in biomedical imaging and machine vision, driving breakthroughs in optical sectioning, super-resolution microscopy, phase imaging, and 3D surface profiling. These applications rely on precise light patterns projected onto samples to extract high-resolution information, where spatial light modulators (SLMs) are widely used for their high speed, precision, and flexibility. However, the limited pixel count of SLMs constrains the spatial bandwidth product (SBP) of structured illumination, creating a tradeoff between the field-of-view (FoV) and spatial resolution. To address this limitation, we develop a mesoscale structured illumination technique that integrates a custom-designed pyramidal lens with an SLM; this technique achieves an unprecedented SBP enhancement of over 100-fold. We demonstrate the generation of mesoscale structured illumination fields containing more than 5000 stripes with a 6-step phase shift using only a 1024-pixel SLM. Leveraging this approach, we further develop a multicolor super-resolution structured illumination microscopy PyramidalSIM, which delivers a reconstructed FoV of 4K pixels and doubled spatial resolution yet utilizes only 609 SLM pixels. Our approach effectively surpasses the intrinsic SBP limitation of SLMs, paving the way for applications such as high-throughput super-resolution imaging and large-scale surface metrology.