Optical analog computing has emerged as a promising solution to overcome the limitations of digital data processing due to its inherent parallelization and energy efficiency. Recent advances in flat optics have leveraged engineered nonlocality for direct analog image processing, allowing for a significant reduction in the footprint of optical systems for compact integration. The engineering of photonic nanostructures to date has focused on employing resonant elements in linear optical systems, resulting in limited bandwidth and a narrow scope of applications. Recently, a group of scientists from Italy and the USA has merged nonlinear phenomena with engineered nonlocality to enable new regimes of image processing. The non-resonant nature of the nonlinear response, characterized by Volterra kernel operation, allows for broadband nonlocal response that surpasses traditional linear regimes. Remarkably, by simplifying the system to a flat sheet with uniform nonlinear susceptibility, they showcase high-contrast edge detection at the second-harmonic frequency through nonlinear spatial filtering. These findings mark important steps towards nonlinear image processing through practically simple systems, thereby unlocking new engineering freedoms for enhanced functionalities and reconfigurability.

---

---

 Add Comment

You must log in to add comments.