In the realm of modern imaging, the demand for versatility is at an all-time high. The ability to capture not just the overall brightness of an object, but also the intricate details of its boundaries, interfaces, and structural transitions, is becoming increasingly crucial. This is especially true in fields like biology, where identifying cell membranes and tracking morphological changes are essential, and in machine vision, where structural cues are vital. Traditionally, achieving both conventional bright-field images and edge-enhanced images has been a complex and resource-intensive task, often requiring multiple optical elements, sequential measurements, or computational post-processing. This not only increases system size and complexity but also slows down data acquisition, which is a significant challenge when dealing with dynamic targets.
Metasurfaces and metalenses have emerged as a promising solution to this problem, offering a way to replace bulky lenses with flat, microstructured devices that can directly shape wavefronts. However, most compact meta-imaging systems still face two significant challenges. Firstly, they often provide a single imaging function at a time, requiring switching mechanisms to alternate between modes, which can be cumbersome and slow. Secondly, they typically operate reliably only over a limited field of view, leading to a noticeable degradation in image quality when the object or illumination becomes off-axis.
A groundbreaking study published in Opto-Electronic Advances introduces a novel solution to these constraints. The research team, comprising experts from Pohang University of Science and Technology (POSTECH), Korea University – Sejong Campus, Harbin Engineering University, and the Shanghai Institute of Optics and Fine Mechanics, has developed a polarization-encoded trichannel metalens. This innovative device performs parallel bright-field imaging and multi-order edge detection within a single, compact, and static meta-optical platform.
The key innovation lies in the integration of three distinct optical operators into three polarization-defined channels. Channel 1 (RCP to LCP) is dedicated to bright-field imaging, Channel 2 (RCP to RCP and LCP to LCP) to first-order edge detection, and Channel 3 (LCP to RCP) to second-order edge detection. This design enables all three imaging modalities to be recorded simultaneously within a single CCD imaging area under linearly polarized illumination and detection, without the need for dynamic switching or additional optical components.
The metalens demonstrates stable and clearly separated parallel imaging performance over a wide field of view ranging from -10° to 10°. Key device parameters include a numerical aperture of 0.503 and a focal length of 430 µm at 635 nm. The measured focusing efficiencies are 41.7% (Channel 1), 25.1% and 28.3% for the two polarization-preserved realizations of Channel 2, and 14.5% (Channel 3).
The metalens was validated on a USAF 1951 resolution target and on representative biological specimens (Spirogyra conjugation and Volvox). The edge-enhanced channels in these demonstrations highlighted structural boundaries and textures that are difficult to discern in conventional bright-field images, showcasing the metalens's ability to provide valuable insights in various fields.
This research not only addresses the immediate challenges in meta-imaging but also contributes to a broader roadmap toward compact meta-optical platforms with increased channel capacity, enriched imaging operators, and improved power allocation and channel isolation. The use of polarization for channel addressing is a particularly simple and effective approach, paving the way for high-throughput imaging platforms where intensity and edge information can be accessed simultaneously.
The corresponding authors, Prof. Junsuk Rho and Prof. Trevon Badloe, are at the forefront of metasurfaces and flat optics research, focusing on polarization-multiplexed meta-imaging, multi-functional metalenses, and compact optical systems that integrate imaging and optical analog information processing. Their work is a testament to the potential of these technologies to revolutionize imaging and edge analysis in various applications, from biomedical imaging to real-time edge analysis and machine vision.
