Multiscale Aperture Synthesis Imager Surpasses Diffraction Limits, Enabling Coherent Imaging Without Overlapping Measurement Regions

Synthetic aperture imaging empowers observations across diverse fields, from radar technology to astronomy, but realising its full potential demands overcoming significant technical hurdles. Ruihai Wang, Qianhao Zhao, and Tianbo Wang, alongside colleagues at their institutions, now present the multiscale aperture synthesis imager (MASI), a system that fundamentally simplifies the complex process of wavefield synchronisation. MASI employs a distributed array of sensors that operate both independently and coherently, effectively bypassing the diffraction limit traditionally imposed on single receivers. By combining propagated wavefields through a novel phase synchronisation scheme, the system eliminates the need for precise overlapping measurement regions, and instead generates naturally expanded, high-contrast visualisations without the use of lenses. This achievement enables the resolution of sub-micron features at ultralong working distances, and reconstructs three-dimensional shapes over fields spanning several centimetres, representing a substantial advance in scalable synthetic aperture imaging.

Multiscale Imaging Breaks Diffraction Limit

This research introduces the multiscale aperture synthesis imager, or MASI, a new approach to optical imaging that overcomes limitations of conventional synthetic aperture methods. MASI utilizes parallelism, dividing complex imaging challenges into manageable sub-problems, and achieves high-resolution imaging without demanding precise synchronization between sensors. The system employs an array of coded sensors that operate independently yet coherently, surpassing the diffraction limit of a single receiver. A novel reconstruction algorithm combines information from these sensors to create a single, high-resolution image, significantly reducing the need for precise calibration.

Lensless Imaging via Fourier Ptychography

This work explores computational imaging and ptychography, powerful techniques that move beyond traditional lens-based imaging. Ptychography reconstructs high-resolution images from a series of diffraction patterns, eliminating the need for lenses and potentially achieving higher resolution. Several variations are investigated, offering unique advantages in performance and efficiency. Lensless imaging offers reduced cost, complexity, and the potential to image in challenging environments. Researchers are adapting ptychography for specific applications and overcoming limitations through high-throughput imaging and three-dimensional reconstruction.

Techniques like integral imaging and light-field imaging capture and reconstruct 3D information, while intensity diffraction tomography reconstructs 3D objects from diffraction patterns. Polarization-sensitive imaging captures and analyzes the polarization of light to reveal material properties, and multi-modal imaging combines ptychography with other imaging modalities like fluorescence microscopy. The use of freeform illuminators and continuous height-varying modulation further optimizes image quality. These technologies have a wide range of potential applications, including biomedical imaging, materials science characterization, forensic science analysis, security surveillance, remote sensing, and microscopy improvements.

Researchers are also developing advanced system-level considerations, including image registration, subpixel alignment, efficient data processing pipelines, system integration, miniaturization, and fiber-based systems. Future research directions include integrating deep learning for improved image reconstruction and analysis, utilizing neural wavefront shaping to correct for distortions, and applying the transport of intensity equation for image reconstruction. Connections to synthetic aperture radar and gigapixel imaging demonstrate the broad applicability of these computational imaging techniques. This work highlights the revolution occurring in optics, moving beyond the limitations of traditional lenses and embracing computational power.

Parallel Sensor Array Achieves High Resolution Imaging

The multiscale aperture synthesis imager, or MASI, represents a significant advancement in imaging technology. This system overcomes limitations of traditional synthetic aperture methods by employing parallelism to address complex imaging challenges. MASI utilizes an array of coded sensors that operate independently yet coherently, surpassing the diffraction limit of a single receiver. A key innovation is a computational phase synchronization scheme that eliminates the need for overlapping measurement regions to establish phase coherence between sensors. Experiments demonstrate that MASI successfully recovers complex wavefield information from individual sensors, even in the presence of low-frequency aberrations.

Conventional phase retrieval methods struggle with slowly varying phase components, but MASI’s coded sensors reliably recover both step phase transitions and linear phase gradients with minimal offset from ground truth. This robust performance stems from coded surface modulation, which converts phase variations into detectable intensity variations, enabling accurate wavefield reconstruction. The system achieves pixel super-resolution reconstruction from intensity-only diffraction measurements, significantly enhancing image detail. A key achievement of MASI is its ability to expand the imaging field without the use of lenses.

Diffraction naturally expands the field of view beyond the physical dimensions of each sensor, effectively eliminating gaps in the final reconstruction. The prototype system, consisting of a 9-sensor array, operates with sub-pixel shifts, enabling complex wavefield recovery. Sensors can be positioned on surfaces at different depths and locations without requiring precise alignment, dramatically simplifying system implementation while maintaining the ability to synthesize a large virtual aperture. This design tolerance allows for scalable, long-baseline optical imaging, similar to distributed telescope arrays used in radio astronomy.

Multiscale Imaging Without Precise Alignment

This research presents the multiscale aperture synthesis imager, or MASI, a new approach to synthetic aperture imaging that overcomes longstanding challenges in achieving coherent measurements across multiple receivers. By employing a distributed array of coded sensors and a novel computational phase synchronization strategy, MASI successfully reconstructs high-resolution images without the need for precise physical alignment or overlapping measurement regions, a significant advancement over traditional interferometric methods. The system demonstrates the ability to resolve sub-micron features at ultralong working distances and reconstruct three-dimensional shapes over centimeter-scale fields, all without the use of lenses. MASI’s architecture shares conceptual similarities with large-scale projects like the Event Horizon Telescope, but replaces complex hardware synchronization with a computationally efficient process that optimizes a single global phase offset per sensor. This allows for flexible sensor placement and scalable imaging, particularly advantageous for long-baseline optical imaging where maintaining interferometric coherence is difficult. While MASI, like sparse aperture systems, faces limitations due to incomplete frequency coverage resulting from gaps between sensors, future work may focus on addressing these limitations and further expanding the capabilities of this new imaging architecture for a wider range of applications.